Onat Şahin, Mohammad Altillawi, George Eskandar +2cs.CV
3D scene representations like NeRF and 3D Gaussian Splatting (3DGS) suffer severe artifacts in sparse-view settings. Recent generative 3D artifact fixers attempt to address this, but rely on paired corrupted and clean renders requiring costly, per-scene reconstructions across varying view configurations. While 2D image augmentations act as instant regularizers, no explicit equivalents exist for 3D representations to preserve spatial consistency across views, an essential property for 3D-aware training. We propose 3D Morphological Perturbations as an optimization-free regularizer that preserves spatial consistency. Leveraging explicit 3DGS, we treat each Gaussian as a fundamental building block - analogous to a 2D pixel - and apply perturbations across its morphological parameter space via scale, rotation, and pruning. Our method eliminates per-scene 3DGS optimization loops from dataset curation while enabling models to learn stronger geometric priors than sparse-view baselines in diagnostic ablations conducted on a lightweight video diffusion sandbox. Scaled to a 14B-parameter video model via ControlNet, our approach maintains visual fidelity while reducing mean depth error by 12.5% over state-of-the-art image-to-image 3D artifact refiners, ultimately boosting downstream robotics policy success rates by up to 8.0% across 3 of 4 manipulation tasks.
Training-free, camera-controlled novel view synthesis from a single image using pre-trained video diffusion models often becomes unstable under large camera motion and long generation horizons. Existing approaches commonly combine several inference-time components, making it unclear which design choices are most important for stability. We show that the main source of stability is simple. Decomposing camera motion into small autoregressive steps limits per-step geometric distortion and reduces error accumulation. A controlled camera-step study shows that performance remains stable for small motions and degrades more strongly as the per-step motion approaches $18$-$20^\circ$. We further evaluate geometry-constrained spatial attention and low-frequency appearance anchoring as supporting refinements, together with an efficient registration-free warping pipeline. Across RealEstate10K and MegaScene, CamTrol++ improves temporal and geometric consistency, downstream 3D reconstruction quality, and generation efficiency over training-free baselines. The method remains effective for 56-frame generation and under substantial controlled depth corruption. These results show that careful control of camera motion at inference time can substantially improve the stability of camera-controlled novel view synthesis without retraining or modifying the diffusion backbone.
Javier Alejandro Lopetegui Gonzalez, Paul Pacaud, Cordelia Schmidcs.CV cs.RO
World Action Models (WAMs) leverage the capabilities of large-scale pretrained video diffusion models to jointly predict future observations and actions, inheriting rich visual and physical priors from internet-scale video. This has made them a promising paradigm for robot policy learning, yet the prevailing models operate exclusively on RGB observations and do not leverage 3D information. To bridge this gap, we introduce a Spatially Aware World Action Model (SA-WAM), which repurposes a pretrained video model for joint action, RGB, and depth prediction, enabling 3D-aware world modeling and action prediction within a single diffusion backbone. We use a nonlinear encoding that maps the unbounded depth signal into the bounded input domain expected by the frozen VAE tokenizer. This allows us to reuse the tokenizer without 3D-specific fine-tuning, incorporating geometric information without sacrificing the pretrained priors. SA-WAM achieves state-of-the-art results on the RoboCasa and LIBERO-Plus benchmarks, while simultaneously improving future-state predictions. Furthermore, SA-WAM outperforms strong baselines in real-world evaluation using a UR5 robotic arm, with strong gains in randomized environments. We analyze the correlation between world model prediction quality and rollout success, providing insights into WAM performance and avenues for its improvement.
Explorable image-to-scene generation is essential for applications in gaming, robotics, and virtual reality. Existing methods based on video diffusion model (VDM) commonly rely on incomplete conditioning signals such as sparse point clouds or 2D panoramas, leading to stochastic hallucinations, long-term drifts and suboptimal 3D consistency. We present SpatialCrafter, a novel two-stage framework that addresses these issues by introducing a global 3D proxy for high-fidelity image-to-scene generation. Specifically, we decompose the generation process into global proxy generation and appearance refinement. For proxy generation, we propose a Point-anchored Sparse Structure~(PaSS) Flow module that predicts a spatially aligned and geometrically consistent 3D proxy. For appearance refinement, we re-frame the VDM as a Generative Deferred Refiner which synthesizes high-frequency photorealistic details upon proxy-defined scene geometry. To better integrate the proxy with the pre-trained VDM, we introduce Parallel Geometry Injection and Proxy-Aware Corruption training strategies, which improve robustness to proxy artifacts without disrupting the pretrained generative manifold. Furthermore, as no suitable dataset exists for this explorable scene generation task, we construct a new large-scale dataset of 115K scenes. To the best of our knowledge, it is the first hybrid dataset for image-to-scene generation. Extensive experiments on both synthetic and real-world datasets show that SpatialCrafter outperforms state-of-the-art methods, mitigates long-term drift, and remains robust and consistent under rapid camera motion and extreme viewpoint changes. Code, models, and the newly constructed dataset will be publicly released. See more at https://fangchuan.github.io/SpatialCrafter/.
3D Gaussian Splatting (3DGS) has achieved remarkable success in novel view synthesis; however, reconstructions under sparse views often exhibit noticeable artifacts. While recent video diffusion models provide strong spatio-temporal priors for 3DGS restoration, directly fine-tuning them for restoration is suboptimal, as they lack awareness of the underlying multi-camera geometry, resulting in multi-view inconsistencies. In this work, we propose a novel 3D-aware video restoration framework designed to enhance the quality of sparse 3DGS reconstruction. Specifically, we construct a large-scale 3DGS video dataset to enable specialized fine-tuning. To bridge the gap between 2D video generation and 3D multi-view constraints, we introduce a camera-conditioned geometric prior. By using the first and last frames as boundary anchors and encoding the corresponding camera relationships, we explicitly inject spatial structure into the video generation pipeline. This boundary-anchored, camera-aware prior guides the network toward geometrically grounded restoration that remains coherent across viewpoints. Extensive experiments show that, among video-prior restoration methods, our approach attains the best pixel- and structure-level fidelity (PSNR/SSIM) and improves multi-view consistency, while remaining competitive in perceptual quality (LPIPS).
We present 4DAnyone, a framework for reconstructing 4D humans from an uncalibrated monocular video by generating reconstruction-grade multiview-consistent videos and lifting them into 4D Gaussian Splatting (4DGS). Existing camera-controlled video diffusion models synthesize plausible novel-view videos but fail to maintain consistency when scaled to the tens of target views required for 4DGS reconstruction. We identify this failure as a bounded-attention-context problem: when target views exceed the capacity of a single DiT forward pass, they must be split into groups, exposing two coupled bottlenecks. On the reference-context side, conditioning on all previously generated views grows as $O(N)$, weakening cross-view appearance guidance. On the target-context side, disjoint groups cannot directly exchange information, causing global structural drift. 4DAnyone addresses both bottlenecks with two complementary designs: Reference Context Packing (RCP) compresses growing reference views into a fixed-length mixed-resolution context with $O(1)$ reference-context complexity, while Target Context Routing (TCR) rotates target-view groupings during denoising to share context across groups at high-noise steps and stabilize details at low-noise steps. We further build the MVGameHuman dataset using our in-house game engine and combine it with light-stage and in-the-wild video datasets for training. Experiments on DNA-Rendering and DyMVHumans show that 4DAnyone outperforms prior methods in both novel-view video quality and downstream 4DGS reconstruction, with robust in-the-wild generalization. See our project page for video results and source code: https://4danyone.github.io.
Driving scene reconstruction and rendering, especially with 3D Gaussian Splatting, has become an important component of autonomous driving simulation. However, rendered views often degrade under extrapolated ego trajectories and scene edits, producing blurry structures, temporal flicker, and foreground-background misalignment. Existing refinement methods are commonly designed for a specific setting, such as image-level novel-view repair or object-editing correction. In this paper, we introduce SPVC, a structured and panoptic video fixing framework for cross-dataset driving scene rendering. The name summarizes four design principles. (1) Structured fixing denotes the use of explicit spatial conditions, including camera pose, 3D bounding boxes, and HD maps, to guide the repair process and reduce uncontrolled hallucination. (2) Panoptic fixing refers to correcting both background rendering artifacts, such as distorted roads, buildings, and lanes, and foreground vehicle artifacts introduced by scene editing, such as inconsistent object appearance. (3) Video fixing means that the model operates on driving sequences rather than isolated frames, allowing temporal cues to be used during artifact correction. (4) Cross-dataset fixing means that a single shared network is trained and applied across multiple driving datasets, reducing the need for dataset-specific or scene-specific fixers. Concretely, we construct paired degraded-clean training data by simulating under-constrained 3DGS rendering and foreground vehicle insertion artifacts, and train a two-stage controllable video diffusion model that first addresses video-level appearance and then refines scene layout with structured controls.
Zepeng Wang, Jiagao Hu, Fuhao Li +3cs.CV cs.AI eess.IV
Videos captured through glass often contain reflections that degrade visual quality and interfere with downstream vision tasks. Although single-image reflection removal has been extensively studied, video reflection removal remains largely underexplored due to the lack of paired video data, temporally coherent removal models, and dedicated evaluation benchmarks. We present a closed-loop framework that unifies physics-grounded reflection simulation, diffusion-based video dereflection, and benchmark evaluation. Our S2R-Synthesis pipeline generates paired reflected and reflection-free videos by performing physics-grounded augmentation in the structure space and rendering realistic reflected videos with a trained video diffusion renderer; the augmentation models key glass-related effects including roughness-induced blur, thickness-induced ghosting, and reflectance variation. Based on the synthesized data, we introduce S2R-Removal, the first diffusion-based video reflection removal model, which adapts a pretrained video diffusion prior through reflection-aware latent adaptation and one-step pixel-geometric refinement, recovering the clean transmission in a single denoising step. We further build S2R-Bench, the first benchmark for video reflection removal, supporting both full-reference evaluation and real-world human perceptual assessment. Experiments on S2R-Bench and multiple public image benchmarks demonstrate state-of-the-art performance and faster inference than even non-diffusion baselines, and validate the effectiveness of S2R-Synthesis. Project page: https://codingwzp.github.io/VideoDereflection_S2R.
4D generation synthesizes dynamic 3D scenes from conditions such as text or images. Existing methods either reconstruct generated RGB videos with a separate 4D model or adapt a particular video generator to predict geometry directly. The former suffers from distribution mismatch and error propagation, whereas the latter ties 4D prediction to a specific generator and may require retraining when the generator or conditioning regime changes. We ask whether the final denoised latents of video models that share a variational autoencoder (VAE) can instead provide a reusable interface to explicit 4D prediction. Building on this insight, we introduce direct latent-to-4D generation and instantiate it as Latent-to-4D, which bypasses RGB by aligning a video latent with the token grid of a pretrained 4D decoder and refining it through frame-wise and global spatiotemporal attention. Trained on roughly 1K existing reconstruction clips, a single checkpoint transfers unchanged across multiple video diffusion transformers within the same VAE family. On Text4D-200 and I4D-200, Latent-to-4D surpasses matched same-latent Wan+4RC cascades in projection-based DINO-F1 by 2.88--3.45 and 5.81 points, respectively, while also being preferred by human raters for geometry, temporal stability, and overall quality.
Given textual task instructions, generating step-by-step visual instructions as an image sequence requires the simultaneous satisfaction of multiple properties, specifically step faithfulness, cross-image consistency, and per-frame visual quality. Existing text-to-image generation approaches rarely meet all three properties, owing to independent sampling that breaks consistency, finetuning on low-quality video that degrades per-frame quality, and frozen backbones that lack multi-step understanding. In this work, we propose InstructionCrafter, a diffusion-based framework with the key idea of separating the optimization of temporal and instructional alignment from per-frame visual quality via (1) spatial-freeze training and (2) instruction-aware adapters. Built on a pretrained video diffusion backbone, InstructionCrafter freezes the spatial layers that control per-frame detail and updates only temporal and text-conditioning pathways to learn instruction semantics and inter-step relations, which preserves the generative prior for per-frame quality and reduces trainable parameters by about 50 percent compared with full finetuning. We also introduce two lightweight adapters that enhance the model's understanding of instructional context. The Consistent Adapter aggregates textual cues from the entire instruction sequence and from neighboring steps to keep object identity and attributes consistent across frames, and the Context-Aware Temporal Adapter converts cross-attention outputs into biases for temporal self-attention, explicitly propagating inter-frame relations. Extensive experiments on two benchmark datasets demonstrate state-of-the-art overall performance on step faithfulness, cross-image consistency, and per-frame visual quality while significantly reducing noise, blur, and spurious subtitles. Our code and trained models will be publicly available.
Controllable surgical world models can provide a generative foundation for surgical artificial intelligence and simulation by synthesizing realistic instrument--tissue interactions. However, existing methods lack a unified multimodal control paradigm, while direct fusion of heterogeneous visual conditions often causes anatomical distortion, instrument appearance drift, and temporally inconsistent interactions. In this work, we propose {Surg-UniWorld}, a unified surgical world model with multimodal control experts. Surg-UniWorld first constructs a {Hierarchical Surgical Anchor} from first-frame appearance and hierarchical semantic masks to preserve persistent scene identity, anatomical organization, and interaction boundaries. {Anchor-Relative Modality Experts} then interpret edge, depth, and optical-flow evidence relative to the shared anchor, capturing complementary boundary, geometric, and motion information. A {Multimodal Control Expert} further performs contribution-preserving stage-wise composition of the activated modality increments and generates control hints for the Wan2.2 video diffusion backbone. To support multimodal surgical world modeling, we further construct Cholec80-SurgWAM, a benchmark for controllable surgical video generation. Extensive experiments demonstrate that Surg-UniWorld consistently outperforms existing controllable video generation methods and surgical world-model baselines in generation quality, temporal consistency, and multimodal controllability.
World action models (WAMs) built on video generation backbones are a rising recipe for robot learning, yet remain confined to tabletop manipulation. Mobile manipulation demands simultaneous locomotion and whole-body manipulation amid scene-scale dynamics, yet is still dominated by dynamics-blind visual encoders with hand-crafted coordination. We bridge this gap with MobileWAM, a mixture-of-transformers architecture that fuses a pretrained video diffusion transformer with a lightweight action expert through layerwise joint attention, translating internet-scale motion priors into whole-body control. To reconcile the heterogeneous dynamics of moving and manipulating, each feed-forward layer of the action expert becomes a three-expert mixture of shared, locomotion, and manipulation experts, softly routed by the motion intent in the action tokens. To densify supervision, we further propose Chain-of-Foresight (CoF): intermediate representations sequentially predict a chain of future latent chunks, each step conditioned on its predecessor. CoF pairs naturally with our decoupled video--action denoising scheme. At deployment, the WAM serves as a pure current-frame encoder; foresight acts only through gradients, so at inference the foresight chain and video generation are discarded, leaving only policy-level cost. MobileWAM surpasses state-of-the-art mobile manipulation policies on ManiSkill-HAB and fine-tunes to a real ARX Lift2 mobile manipulator across diverse tasks with strong generalization. Code will be released soon.
Video diffusion transformers (vDiTs) generate high quality but pay quadratic self-attention cost, making inference prohibitive at video-token scales. The challenge is input-adaptive sparsity: selecting critical Q/K/V tokens with negligible overhead and executing them for end-to-end gains. We present SPADE, a training-free sparse-attention engine of three parts: (i) vDiT-SSR, a specification defining 3D blocking candidates and formalizing dynamic masks via Summarizer/Estimator expressions; (ii) runtime scheme generation using SICS and a head-wise policy; and (iii) an executor with low-overhead index search, flash block-sparse attention, and kernel grouping. Across Hunyuan-Video and Wan 2.1/2.2 for text-to-video and image-to-video generation, SPADE raises sparsity and speed while preserving quality, accelerating attention by 2.26x-3.40x and end-to-end inference by 1.49x-1.80x. Our code is open-sourced at https://github.com/6somehow/DAC-SPADE.
Accurate prediction of object trajectories during manipulation is essential for closing the perception-action loop. Progress is limited on two fronts: available datasets lack fine-grained language-to-motion annotations, and existing predictors either rely on privileged inputs such as video, depth, or CAD models, or recover motion from fully generated videos through costly, error-prone perception pipelines. We close the supervision gap with the MOVE dataset, 5,038 object-centric egocentric trajectories, each paired with a fine-grained natural-language instruction rather than a coarse verb-noun label. We further propose DreamTraj, which predicts a 6-DoF object trajectory from a single RGB image and a task instruction, requiring no video, depth, or CAD model at inference: rather than generating a video, it reads motion from the internal representations of a frozen image-to-video diffusion model at an early denoising step. A lightweight flow-matching Reader decodes query-key attention tracks and pooled hidden states into relative 6-DoF poses. To our knowledge, this is the first approach to directly decode object 6-DoF trajectories from intermediate video diffusion representations rather than generated pixels. DreamTraj sets a new state of the art on both translation and rotation against forecasters that consume multi-frame or privileged inputs, and runs 4.6x faster than generate-then-extract pipelines.
Novel view rendering of large and complex reconstructed scenes is becoming increasingly photorealistic. However, most reconstructions remain static and lack the ambient motion that makes environments immersive. We present AniGS, a method for scene-level animation of 3D Gaussian Splatting (3DGS) reconstructions that adds subtle, distributed dynamics, e.g., vegetation motion, while preserving rigid structures. Unlike existing 3D animation techniques which are limited to object-centric subjects or small regions, AniGS is designed for large, cluttered, navigable scenes. AniGS represents the scene with a canonical 3DGS and models motion using a time-conditioned deformation field. To animate the entire scene, we leverage a pretrained video diffusion model and introduce an iterative dataset--model update strategy that progressively expands viewpoint coverage and repeatedly updates camera-fixed training videos using a render-and-refine scheme. To prevent artifacts from unintended motion in static areas, we further introduce a composed video-to-video refinement scheme that restricts motion to desired regions. Experiments on five real-world, large-scale outdoor scenes demonstrate that AniGS produces natural ambient dynamics and high-quality novel view videos, enabling more immersive viewing experiences of reconstructed environments.
High-resolution video diffusion models built on Diffusion Transformers (DiTs) deliver strong fidelity but quickly exhaust the memory budget of a single workstation. A 100 billion-plus parameter DiT easily requires over a terabyte of persistent state, while naive spatiotemporal self-attention grows quadratically in sequence length. These two walls -- parameter memory and activation memory -- prevent researchers from adapting massive generative models without large GPU clusters. We revisit this problem from a systems perspective and introduce MegaSlide-DiT, a prototype that demonstrates how a pre-trained 105B DiT can be adapted on a single H200 GPU with 1.5 TB of host RAM. Our key insight is that the GPU need not own the model state: all persistent weights, master weights and optimizer moments remain in host memory, while only transient shards are streamed to the GPU on demand. Simultaneously, we replace quadratic global attention with 3D Deformable Slide Attention (3D-DSA), a motion-adaptive local attention operator that reduces both memory and computational complexity to linear in the sequence length. We report detailed memory accounting, execution traces and evaluation results to substantiate our design. MegaSlide-DiT does not claim to train a 105B model from scratch on a single GPU, nor does it magically solve bandwidth limits; rather, it offers a pragmatic path for full-parameter adaptation of massive video diffusion models on high-end workstations.
Thang-Anh-Quan Nguyen, Moussab Bennehar, Luis Guillermo Roldao Jimenez +5cs.CV
Reliable perception under diverse weather conditions remains a major challenge for autonomous driving systems. A common strategy to improve robustness is either to synthesize adverse weather conditions for training perception models or to apply weather-removal techniques to recover clean inputs. However, existing approaches typically rely on synthetic data augmentation or physics-based, task-specific models that require paired training data and often struggle to generate realistic weather effects or generalize robustly to out-of-domain scenarios. Toward this problem, we present Cyclone, a unified framework for weather editing based on latent diffusion, equipped with cycle-consistent constraints and knowledge from image-text models. Cyclone enables the generation of multiple weather conditions across diverse scenes while eliminating the need for paired data. Experimental results show that our approach produces more realistic, structure-preserving outputs than existing baselines and leads to consistent improvements across several downstream driving perception tasks. Furthermore, we demonstrate that Cyclone can be distilled to a video diffusion model for temporally consistent weather editing.
Wenxuan Miao, Haosong Liu, Weiming Hu +9cs.AR cs.AI
Video diffusion transformers (vDiTs) generate high quality video but introduce extremely high compute cost due to the long diffusion timesteps and self attention computation. As diffusion timesteps are reduced, the computation cost of self attention becomes the dominant bottleneck. Existing acceleration approaches largely inherit sparse attention techniques from large language models, which fail to consider the unique spatiotemporal correlation of video data. This paper presents Kaleido, an algorithm hardware codesign that accelerates all operations in vDiTs by exploiting channel-wise spatiotemporal correlations in latent space. Based on this insight, we propose a lightweight channelwise reuse algorithm that skips redundant computations by reusing partial results while preserving higher generative quality than prior methods (>17 dB). To efficiently support this algorithm, we design a systolic array like accelerator with reconfigurable processing elements and a lightweight data dispatcher to mitigate irregular sparsity and data access patterns introduced by our reuse algorithm. Evaluations across three mainstream vDiT models show that Kaleido achieves up to 5.9x speedup and 16.0x energy savings over state of the art accelerators.
Zixin Guo, Yehonathan Litman, Yifeng He +3cs.CV cs.GR
Video relighting requires balancing long-form temporal consistency with a physically grounded understanding of light transport, which depends on accurate estimation of intrinsic scene properties such as materials, geometry, and illumination. Existing methods follow two paradigms: (1) reconstruct a video's photometric properties via inverse rendering and relight them to a target illumination via forward rendering, using physically-based rendering (PBR) or a neural renderer; these suffer from noisy reconstructions and struggle with hard-to-model effects such as global illumination. (2) Frame the task as generative video-to-video translation conditioned on relighting targets (a target environment map or text); this limits relighting control and temporal stability, since diffusion models struggle to translate long-form videos, and is constrained by the availability of input/relit training pairs. We propose LightCrafter, a hybrid pipeline that reformulates video relighting as video translation of a proxy video: rather than translating the input video directly to the target, we translate a PBR rendering of the input under the target illumination to the final target. This bakes illumination targets into the PBR proxy, removing the need to teach the diffusion model illumination concepts like environment maps, and enables more intricate lighting control while naturally providing long-form temporal consistency. We show PBR renders alone already outperform some prior art but struggle with effects like global illumination; to capture these, we leverage photometric priors in video generation models by post-training CogVideoX on synthetic video pairs and real-world unpaired videos. We outperform prior state-of-the-art on existing real-world relighting benchmarks and contribute a synthetic benchmark for further analysis. We will release our dataset, benchmark, metrics, and code.
Earth Observation (EO) forecasting aims to predict future Earth surface dynamics from satellite observations under changing meteorological conditions. In this paper, we view this task as a partially observed, weather-driven world modeling problem, in which weather acts as a conditioning signal, while forecasting remains uncertain due to sparse observations and unobserved land-surface states. However, existing methods do not fully capture this setting: deterministic models collapse uncertainty into a single future prediction, while diffusion-based methods typically treat weather variables as undifferentiated conditioning signals, and existing benchmarks focus mainly on reconstruction accuracy rather than whether forecasts respond correctly to changed weather forcing.We introduce EO-WM, a video diffusion transformer for multispectral EO forecasting. EO-WM incorporates a physically informed conditioning framework that represents meteorological forcing through a climatological baseline, weather anomalies, and cumulative physical stress signals. Specifically, it separates baseline and anomaly through distinct conditioning pathways, and accumulates anomalous forcing over time to capture sustained heat and drought stress. To evaluate weather-response behavior beyond standard metrics, we introduce two diagnostic benchmarks: an Extreme Summer Benchmark for severity-aware prediction of vegetation degradation under extreme weather, and a Seasonal Matched-Pair Benchmark for testing response fidelity under changed weather forcing. Experiments show that EO-WM reduces the error in predicted Normalized Difference Vegetation Index (NDVI) decline amplitude by a relative 5.63% and improves directional hit rate by a relative 7.80%, while remaining competitive on standard pixel-level metrics. The benchmarks and model will be made open-source at https://github.com/Luo-Z13/EO-WM.
Synthesizing a novel-view video from a monocular reference video along a target camera trajectory requires both geometric consistency and motion fidelity with respect to the reference video. Existing methods based on explicit 3D representations are limited by the accuracy of off-the-shelf reconstruction modules, which often produce inaccurate geometry for dynamic objects in monocular videos. In contrast, camera-conditioning-only methods can achieve high visual quality but often struggle to preserve geometric and motion consistency. In this work, we introduce MVTrack4Gen (Multi-View point Tracking for Novel-View Generation), a motion-aware training framework that leverages multi-view point tracking as an additional geometric and motion supervision signal for camera-conditioning-only novel-view video diffusion models. Our key finding is that specific attention layers encode strong correspondence cues, where query features attend to key features at geometrically corresponding locations across views and over time, and the misalignment of these correspondences causes motion inconsistency. Based on this observation, we route these features into an auxiliary multi-view tracking head and jointly train the diffusion model with a point-tracking objective. By explicitly strengthening these motion-aware correspondences, MVTrack4Gen improves existing models to better follow the motion in the reference view and maintain cross-view geometric consistency. Across diverse benchmarks, our method achieves state-of-the-art geometric consistency and competitive camera accuracy.
Generating explorable 3D scenes from a single image requires strong generative priors and accurate geometric representations suitable for downstream use. Current video diffusion models offer high-quality generation and implicitly encode multi-view geometric structure in latent space. However, existing feedforward latent scene decoders typically output volumetric 3D Gaussians that lack a well-defined surface, limiting their use in simulation or standard graphics pipelines. This motivates decoding surface-aligned primitives that are not only renderable but also closer to explicit geometric assets. We ask whether compressed video diffusion latents can be mapped directly to explicit surface primitives in a single pass. To this end, we introduce FLAT and, for the first time, show that triangle splats can be decoded directly from video diffusion latents. Compared with decoding 3D Gaussians, predicting flat primitives is notoriously more challenging due to high sensitivity to primitive orientations, oftentimes leading to poor gradient flow. FLAT solves with two key ingredients: a ray-centered rotation parameterization for triangle regression and a novel product window function that improves gradient flow during differentiable triangle rendering. On standard benchmarks, FLAT achieves significantly better geometric accuracy while maintaining competitive visual quality compared to state-of-the-art feedforward baselines. We further show that a lightweight test-time refinement step converts the predicted triangle soup into a fully opaque, game-engine-ready representation that supports real-time rendering. By evaluating 3DGS, 2DGS, and triangle splatting variants under an identical training setup, we provide the first systematic analysis of representation tradeoffs in feedforward scene generation. The project page is available at https://flat-splat.github.io
Underwater 4D reconstruction remains challenging due to the coupling between degraded light transport in participating media and dynamic water variations. Most existing Methods are developed under in-air assumptions and do not explicitly account for underwater absorption and backscatter. Additionally, near-static assumptions make these approaches sensitive to drifting particles and dynamic distractors , leading to unstable geometry and inconsistent cross-view results. To address these issues, we propose a generative framework for underwater 4D reconstruction, named Ocean4D, which is built on two complementary components. Specifically, 4D-GCC constructs 4D geometrically consistent conditioning with improved cross-frame coverage, while the Medium-Aware Block performs implicit medium-aware denoising in the latent diffusion process to stabilize underwater appearance under absorption and scattering. Given a monocular video and target cameras, our method generates videos along the target trajectories while preserving global structure and cross-view consistency. Extensive experiments on both dynamic and static underwater benchmarks demonstrate state-of-the-art performance on underwater reconstruction.
Kaijian Wang, Yuanyuan Xu, Fanjiang Ye +5cs.DC cs.LG
Video diffusion has quickly grown into a key generative serving workload, yet producing each clip demands many denoising iterations over large spatio-temporal latents, which puts low-latency inference out of reach on a single device. A denoising step is therefore typically distributed across multiple accelerators, and TPU sub-slices have become an attractive and practical fabric for doing so. Current auto-parallel systems, however, search almost exclusively over logical device meshes and disregard how a chosen sharding is actually laid out on the physical TPU interconnect -- an oversight that leaves large, topology-dependent performance on the table. We address this gap with AoiZora, a compiler-mediated topology planner built for low-latency video diffusion inference on TPU sub-slices. Its guiding principle is to reconnect logical sharding with physical placement by drawing on different points in the compilation flow: AoiZora first eliminates weak sharding candidates from inexpensive pre-compilation IRs, then compiles only the ones that survive and orders their physical placements using compiled HLO together with a topology-aware communication model. The winning plan is realized along the ordinary compiler path, leaving model code, compiler lowering, collective kernels, and network routing entirely intact. On TPU v5e sub-slices, AoiZora reduces Wan 2.1 one-step denoising latency by as much as 1.42x relative to existing solutions.
4D generation (\textit{i.e.}, dynamic 3D generation) has recently emerged as a rapidly growing research frontier due to its powerful spatiotemporal modeling capabilities. However, despite notable advances, existing approaches typically fail to capture the underlying physical principles, producing results that are both physically inconsistent and visually implausible. To overcome this limitation, we present CP4D, a novel paradigm for photorealistic 4D scene synthesis with faithful adherence to complex physical dynamics. Drawing inspiration from the compositional nature of real-world scenes, where immutable static backgrounds coexist with dynamic, physically plausible foregrounds, CP4D reformulates 4D generation as the integration of a static 3D environment with physically grounded dynamic objects. On this basis, our framework follows a three-stage pipeline: \textbf{1)} Firstly, we leverage pre-trained expert models to generate high-fidelity 3D representations of the environment and foreground objects respectively. \textbf{2)} Subsequently, to produce physically plausible trajectories and realistic interactions for these objects, we propose a hybrid motion synthesis strategy that integrates priors from physical simulators with the common sense embedded in video diffusion models. \textbf{3)} Finally, we develop an automated composition mechanism that seamlessly fuses the static environment and dynamic objects into coherent, physically consistent 4D scenes. Extensive experiments demonstrate that CP4D can generate explorable and interactive 4D scenes with high visual fidelity, strong physical plausibility, and fine-grained controllability, significantly outperforming existing methods. The project page: https://anonymous.4open.science/w/CP4D/.
Estimating hand-surface contact pressure from an egocentric view is crucial for AR/VR devices, robotic imitation, and ergonomic analysis. Existing methods often discretize pressure signal and process frames independently, leading to quantization errors and temporal inconsistencies. We present \emph{EgoPressDiff}, a conditional video diffusion framework that generates UV-pressure maps from visual input. The core of our approach is a multi-modal conditioning strategy, introducing a PoseNet and a Vertex Encoder to efficiently extract features from hand pose and 3D mesh vertices. These signals, along with depth information, guide the generative process to ensure the pressure fields are physically grounded. To effectively fuse these heterogeneous features, we further propose a Distribution-Calibrated Spatial Layer, which aligns their statistical properties before combination. Evaluated on the EgoPressure ego-view setting, EgoPressDiff achieves state-of-the-art results, improving Volumetric IoU by over 34\% relative to prior baseline, while reducing MAE and maintaining high temporal accuracy. Our project page is at https://egopressdiff.github.io/.
Junjie Ye, Rong Xue, Basile Van Hoorick +6cs.RO cs.CV
Scaling robot learning requires large-scale, diverse demonstrations, yet real-world data collection via teleoperation remains prohibitively expensive and time-consuming. While video diffusion models offer a promising avenue for data scaling, existing generative approaches are often limited to superficial visual augmentation, or suffer from embodiment hallucinations that yield physically infeasible motions. We present a generalizable embodiment-centric world model that achieves scalable data generation by synthesizing photorealistic demonstrations with novel objects, in novel scenes, and from novel viewpoints. Our approach anchors generation to rendered robot motion while conditioning on explicit scene and object priors, effectively decoupling trajectory execution from environment synthesis. This formulation has the potential to unlock two powerful data scaling capabilities: (1) retrieval and rebirth, which repurposes existing trajectories into entirely new contexts without new motion data; and (2) prop-free teleoperation, where operators manipulate empty air and the model hallucinates the target objects and scene afterwards, eliminating reset time. We demonstrate with real-world experiments that our generated data consistently improves downstream policy performance and significantly reduces real-world data requirements across diverse manipulation tasks.
To study the ability to infer physical dynamics from videos and extrapolate them forward in time, we assemble a dataset of 2D Material Point Method (MPM) physical simulations covering rich physical phenomena such as deformable objects, fluids, kinetic objects, and emitters. We study code generation and video diffusion approaches on this dataset, identifying their strengths and weaknesses by varying the amount of physically relevant side information. The code generation model, beyond giving a working demonstration of automatic synthesis of MPM simulations, reveals that such an approach struggles with inferring physical parameters from visual input, but relative to video diffusion, produces physically and temporally stable extrapolations forward in time, while the video diffusion model more strongly identifies geometric properties from visual input but produces physically implausible extrapolations.
We propose X-WAM, a Unified 4D World Model that unifies real-time robotic action execution and high-fidelity 4D world synthesis (video + 3D reconstruction) in a single framework, addressing the critical limitations of prior unified world models (e.g., UWM) that only model 2D pixel-space and fail to balance action efficiency and world modeling quality. To leverage the strong visual priors of pretrained video diffusion models, X-WAM imagines the future world by predicting multi-view RGB-D videos, and obtains spatial information efficiently through a lightweight structural adaptation: replicating the final few blocks of the pretrained Diffusion Transformer into a dedicated depth prediction branch for the reconstruction of future spatial information. Moreover, we propose Asynchronous Noise Sampling (ANS) to jointly optimize generation quality and action decoding efficiency. ANS applies a specialized asynchronous denoising schedule during inference, which rapidly decodes actions with fewer steps to enable efficient real-time execution, while dedicating the full sequence of steps to generate high-fidelity video. Rather than entirely decoupling the timesteps during training, ANS samples from their joint distribution to align with the inference distribution. Pretrained on over 5,800 hours of robotic data, X-WAM achieves 79.2% and 90.7% average success rate on RoboCasa and RoboTwin 2.0 benchmarks, while producing high-fidelity 4D reconstruction and generation surpassing existing methods in both visual and geometric metrics.