{"total":29,"items":[{"citing_arxiv_id":"2606.30608","ref_index":10,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"UnfoldArt: Zero-Shot Recovery of Full Articulated 3D Objects from Text or Image","primary_cat":"cs.CV","submitted_at":"2026-06-29T17:44:53+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"UnfoldArt uses a two-round structured debate between high-level semantic agents and low-level parameter agents, grounded in generated video, to infer articulation and reconstruct full articulated 3D objects including occluded geometry from text or image inputs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.28128","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"PhysisForcing: Physics Reinforced World Simulator for Robotic Manipulation","primary_cat":"cs.CV","submitted_at":"2026-06-26T14:30:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"PhysisForcing applies trajectory and relational alignment losses to DiT features in video models, improving physical plausibility on R-Bench, PAI-Bench, and EZS-Bench while raising closed-loop robotic success rates from 16% to 24%.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.25344","ref_index":20,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Follow Your Track: Precise Skeleton Animation Controlled by 3D Trajectories","primary_cat":"cs.CV","submitted_at":"2026-06-24T03:18:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ACT is a trajectory-conditioned framework for topology-general skeletal animation that injects 3D point trajectories from monocular video into skeletons via a Routed Trajectory Injector for improved fidelity and temporal consistency.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.18558","ref_index":37,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction","primary_cat":"cs.CV","submitted_at":"2026-06-17T00:19:00+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Introduces a new task of goal-conditioned 3D point motion forecasting along with a 1.16M-video dataset, a 111-category benchmark, and a model that outperforms baselines while transferring to robotics and video generation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.10988","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"AnimaSpark: A Feed-Forward Method for Animating Arbitrary 3D Objects","primary_cat":"cs.CV","submitted_at":"2026-06-09T15:25:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"AnimaSpark is a feed-forward pipeline that renders rigged 3D models to multi-layer images, generates video motion, tracks projected keypoints, and lifts 2D planar transforms into 3D skeletal animation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.04593","ref_index":41,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"4D Reconstruction from Sparse Dynamic Cameras","primary_cat":"cs.CV","submitted_at":"2026-06-03T08:31:43+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Presents a 3D track initialization method, depth-ordering regularization, and batch sampling for 4D reconstruction from sparse dynamic cameras, plus the LetCamsGo dataset showing gains in dynamic regions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.04480","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"IMPose: Interactive Multi-person Pose Estimation with Dynamic Correction Propagation","primary_cat":"cs.CV","submitted_at":"2026-06-03T05:55:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"IMPose introduces dual-level (keypoint and instance) correction propagation with a trajectory bank to turn sparse annotations into dense multi-person pose trajectories in videos.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2606.00321","ref_index":8,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Training-Free Object-Agnostic Jam Detection in Fulfillment Centers","primary_cat":"cs.CV","submitted_at":"2026-05-29T19:54:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A training-free jam detection method monitors persistent occlusion of uniformly sampled reference points and reports 100% precision and 93.33% F1 on 1,069 videos.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.24630","ref_index":6,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DexSIM: Real-time Dexterous Simulation with Unified Causal Video Diffusion","primary_cat":"cs.CV","submitted_at":"2026-05-23T15:39:38+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DexSIM is a bi-directional video diffusion model with hand trajectory embedding and spatial memory cache for real-time dexterous hand-object simulation at 15 FPS.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23856","ref_index":57,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Point Tracking Improves World Action Models","primary_cat":"cs.RO","submitted_at":"2026-05-22T17:08:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"MODERATE","novelty_score":7.0,"formal_verification":"none","one_line_summary":"JOPAT jointly models pixels, point tracks, and actions in a diffusion transformer and reports gains over pixel-only baselines on long-horizon robot tasks with occlusion and off-screen motion.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.23045","ref_index":28,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"The TIME Machine: On The Power of Motion for Efficient Perception","primary_cat":"cs.CV","submitted_at":"2026-05-21T21:22:42+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TIME is a motion-based embedding from point tracks, trained only on synthetic data via masked autoencoding, that matches state-of-the-art video model performance with up to 10,000x less training data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.18743","ref_index":26,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"WorldString: Actionable World Representation","primary_cat":"cs.AI","submitted_at":"2026-05-18T17:58:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Proposes WorldString, a differentiable neural model for the state manifold of actionable physical objects learned directly from 3D or video data as a building block for world models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.15185","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Quantitative Video World Model Evaluation for Geometric-Consistency","primary_cat":"cs.CV","submitted_at":"2026-05-14T17:59:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PDI-Bench computes 3D projective residuals from segmented and tracked points to quantify geometric inconsistency in AI-generated videos.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.14645","ref_index":14,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Vision-Based Water Level and Flow Estimation","primary_cat":"cs.CV","submitted_at":"2026-05-14T10:01:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"An integrated framework is proposed that combines SOTA vision models with statistical modeling and physical priors to improve accuracy in water level detection and flow estimation.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.09538","ref_index":2,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"PhysHanDI: Physics-Based Reconstruction of Hand-Deformable Object Interactions","primary_cat":"cs.CV","submitted_at":"2026-05-10T13:51:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"PhysHanDI achieves full 3D hand and non-rigid object reconstruction by simulating object deformations from hand-induced forces and refining hand models via inverse physics, outperforming prior methods in reconstruction and prediction.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05126","ref_index":29,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ConsisVLA-4D: Advancing Spatiotemporal Consistency in Efficient 3D-Perception and 4D-Reasoning for Robotic Manipulation","primary_cat":"cs.RO","submitted_at":"2026-05-06T16:55:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"ConsisVLA-4D adds cross-view semantic alignment, cross-object geometric fusion, and cross-scene dynamic reasoning to VLA models, delivering 21.6% and 41.5% gains plus 2.3x and 2.4x speedups on LIBERO and real-world tasks.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"where causal attention is applied betweenz geo-3D l and zagg-3D l , and bidirectional attention is applied within each. Figure 4.Efficient 4D-Reasoning.IK (implicit knowledge). Cross-Scene Thinker with Spatiotemporal Consistency Attention (SC-Attn) ensures:1)Three sets of initialized dynamic tokens de- code dynamic object representations for one view (CoTracker [29, 30] supervision), guided by object features from different views; 2)One set of initialized depth tokens decodes global depth for three views (Depth-Anything [75, 76] supervision), guided by multi-view geometric relations;3)These predictions serve as in- termediate visual reasoning for parallel action decoding. Finally, we use only the final layer output,z agg-3D"},{"citing_arxiv_id":"2604.22160","ref_index":31,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"GenMatter: Perceiving Physical Objects with Generative Matter Models","primary_cat":"cs.CV","submitted_at":"2026-04-24T02:18:18+00:00","verdict":"CONDITIONAL","verdict_confidence":"MODERATE","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A hierarchical probabilistic model with parallelized Gibbs sampling segments moving matter across random-dot, camouflaged-texture, and naturalistic-video domains, matching supervised baselines and human perceptual judgments.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.22102","ref_index":9,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Wiggle and Go! System Identification for Zero-Shot Dynamic Rope Manipulation","primary_cat":"cs.RO","submitted_at":"2026-04-23T22:17:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Wiggle and Go! uses system identification from rope motion observations to predict parameters that enable zero-shot goal-conditioned dynamic manipulation, achieving 3.55 cm accuracy on 3D target striking versus 15.34 cm without parameter information.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20841","ref_index":16,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DeVI: Physics-based Dexterous Human-Object Interaction via Synthetic Video Imitation","primary_cat":"cs.CV","submitted_at":"2026-04-22T17:59:55+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"DeVI enables zero-shot physically plausible dexterous control by imitating synthetic videos via a hybrid 3D-human plus 2D-object tracking reward.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.19720","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"ReImagine: Rethinking Controllable High-Quality Human Video Generation via Image-First Synthesis","primary_cat":"cs.CV","submitted_at":"2026-04-21T17:47:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"ReImagine decouples human appearance from temporal consistency via pretrained image backbones, SMPL-X motion guidance, and training-free video diffusion refinement to generate high-quality controllable videos.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"this metric is misleading for video generation: it produces high-quality individ- ualframesbutlackstemporalcoherence,resultinginamuchworseFVDof1.442. Human4DiT also struggles to generalize to the complex clothing and appearance variations in our evaluation set. 4.5 Ablation Study Temporal Consistency Ablation via Tracking Visualization.We compare tem- poral consistency strategies using tracking visualization with CoTracker3 [18] (Tab. 2), including Image-First generation without refinement, low-noise re- denoising only, re-denoising with spatiotemporal spectral regularization, and re-denoising with median filtering. Without temporal refinement, trajectories are fragmented and unstable, par- ticularly in fast-moving regions such as arms (red). Re-denoising alleviates some"},{"citing_arxiv_id":"2604.10415","ref_index":16,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Point2Pose: Occlusion-Recovering 6D Pose Tracking and 3D Reconstruction for Multiple Unknown Objects Via 2D Point Trackers","primary_cat":"cs.CV","submitted_at":"2026-04-12T02:17:49+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A model-free system uses 2D point trackers to achieve causal 6D pose tracking and incremental 3D reconstruction for multiple unseen rigid objects from RGB-D video, with recovery from complete occlusions.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.02586","ref_index":4,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"TrackerSplat: Exploiting Point Tracking for Fast and Robust Dynamic 3D Gaussians Reconstruction","primary_cat":"cs.CV","submitted_at":"2026-04-02T23:43:55+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TrackerSplat pre-positions 3D Gaussians using point tracking trajectories to handle large inter-frame displacements in dynamic scene reconstruction.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2511.18373","ref_index":23,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"MASS: Motion-Aware Spatial-Temporal Grounding for Physics Reasoning and Comprehension in Vision-Language Models","primary_cat":"cs.CV","submitted_at":"2025-11-23T09:43:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"MASS adds spatiotemporal motion signals and 3D grounding to VLMs and releases MASS-Bench, yielding physics-reasoning performance within 2% of Gemini-2.5-Flash after reinforcement fine-tuning.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2511.02830","ref_index":10,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Densemarks: Learning Canonical Embeddings for Human Heads Images via Point Tracks","primary_cat":"cs.CV","submitted_at":"2025-11-04T18:58:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DenseMarks learns a canonical 3D embedding space for human head images by training a Vision Transformer with contrastive loss on pairwise point tracks from in-the-wild videos, plus landmark and segmentation supervision.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2510.18822","ref_index":33,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SAM 2++: Tracking Anything at Any Granularity","primary_cat":"cs.CV","submitted_at":"2025-10-21T17:20:15+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SAM 2++ unifies video tracking across mask, box, and point granularities via task-specific prompts, a unified decoder, task-adaptive memory, and a new multi-granularity dataset, reporting state-of-the-art results.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.22699","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Image-Guided Shape-from-Template Using Mesh Inextensibility Constraints","primary_cat":"cs.CV","submitted_at":"2025-07-30T14:09:39+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"An unsupervised SfT approach using image observations and mesh inextensibility constraints reconstructs deforming 3D shapes 400x faster than prior unsupervised methods while handling severe occlusions better.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2507.04447","ref_index":68,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DreamVLA: A Vision-Language-Action Model Dreamed with Comprehensive World Knowledge","primary_cat":"cs.CV","submitted_at":"2025-07-06T16:14:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DreamVLA uses dynamic-region-guided world knowledge prediction, block-wise attention to disentangle information types, and a diffusion transformer for actions, reaching 76.7% success on real robot tasks and 4.44 average length on CALVIN ABC-D.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"WristviewobservationWrist viewdynamicarea T=0T=10T=20T=30 Figure 3: Visualization of dynamic regions over time. We show the static camera (left) and wrist-mounted camera (right) observations alongside the corresponding dynamic masks generated by our method at multiple time steps. The masks highlight dynamic regions by leveraging optical flow trajectories extracted via CoTracker [68, 67]. Compared to the original observations, our method effectively suppresses irrelevant background and focuses on interaction-relevant areas (e.g., moving objects and end-effector), enabling more structured and efficient action reasoning. queries designated as <dream> and <action> are appended to these multimodal embeddings, where <dream> contains three subqueries (dynamic, depth and semantics), which could be used for the"},{"citing_arxiv_id":"2507.00990","ref_index":53,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Robotic Manipulation by Imitating Generated Videos Without Physical Demonstrations","primary_cat":"cs.RO","submitted_at":"2025-07-01T17:39:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RIGVid shows that filtered AI-generated videos can serve as effective supervision for complex robotic manipulation tasks without any real demonstrations.","context_count":1,"top_context_role":"baseline","top_context_polarity":"baseline","context_text":"point-based tracking, closes part of the gap but consistently struggles when large portions of the object become untrack- able. In contrast, RIGVid's use of a structured 6D object pose trajectory enables robust execution across all tasks, ac- counting for the 17.5% improvement over Gen2Act. This advantage persists when more powerful tracking models like CoTracker3 [53] are used, as shown in App. G. Looking at the task-wise breakdown in Fig. 9, RIGVid maintains high success rates even as object depth varies significantly (such as in the lifting task) or when the ob- jects are thin, small, or partially occluded (such as in plac- ing a spatula or sweeping trash). Other methods frequently struggle in these settings, often failing to recover accurate"},{"citing_arxiv_id":"2504.18576","ref_index":36,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"DriVerse: Navigation World Model for Driving Simulation via Multimodal Trajectory Prompting and Motion Alignment","primary_cat":"cs.RO","submitted_at":"2025-04-22T20:58:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"DriVerse is a generative model that simulates driving scenes from an image and trajectory using multimodal prompting and motion alignment, achieving better performance on nuScenes and Waymo datasets with minimal training.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}