TACO: Taming Diffusion for in-the-wild Video Amodal Completion

• URL: http://arxiv.org/abs/2503.12049v1
• Date: Sat, 15 Mar 2025 08:47:45 GMT
• Title: TACO: Taming Diffusion for in-the-wild Video Amodal Completion
• Authors: Ruijie Lu, Yixin Chen, Yu Liu, Jiaxiang Tang, Junfeng Ni, Diwen Wan, Gang Zeng, Siyuan Huang,
• Abstract summary: This paper tackles the task of Video Amodal Completion (VAC), which aims to generate the complete object consistently throughout the video.<n>We propose a conditional diffusion model, TACO, that repurposes pre-trained video diffusion models.<n>We demonstrate TACO's versatility on a wide range of in-the-wild videos from Internet, as well as on diverse, unseen datasets commonly used in autonomous driving.
• Score: 32.474824991167424
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Humans can infer complete shapes and appearances of objects from limited visual cues, relying on extensive prior knowledge of the physical world. However, completing partially observable objects while ensuring consistency across video frames remains challenging for existing models, especially for unstructured, in-the-wild videos. This paper tackles the task of Video Amodal Completion (VAC), which aims to generate the complete object consistently throughout the video given a visual prompt specifying the object of interest. Leveraging the rich, consistent manifolds learned by pre-trained video diffusion models, we propose a conditional diffusion model, TACO, that repurposes these manifolds for VAC. To enable its effective and robust generalization to challenging in-the-wild scenarios, we curate a large-scale synthetic dataset with multiple difficulty levels by systematically imposing occlusions onto un-occluded videos. Building on this, we devise a progressive fine-tuning paradigm that starts with simpler recovery tasks and gradually advances to more complex ones. We demonstrate TACO's versatility on a wide range of in-the-wild videos from Internet, as well as on diverse, unseen datasets commonly used in autonomous driving, robotic manipulation, and scene understanding. Moreover, we show that TACO can be effectively applied to various downstream tasks like object reconstruction and pose estimation, highlighting its potential to facilitate physical world understanding and reasoning. Our project page is available at https://jason-aplp.github.io/TACO.

Related papers

• MVR: Multi-view Video Reward Shaping for Reinforcement Learning [17.20077949643041]
  Multi-View Video Reward Shaping (MVR) is a framework that models the relevance of states regarding the target task using videos captured from multiple viewpoints.<n>MVR learns a state relevance function that mitigates the bias towards specific static poses inherent in image-based methods.<n>We introduce a state-dependent reward shaping formulation that integrates task-specific rewards and VLM-based guidance.
  arXiv  Detail & Related papers  (2026-03-02T10:24:04Z)
• Can World Models Benefit VLMs for World Dynamics? [59.73433292793044]
  We investigate the capabilities when world model priors are transferred into Vision-Language Models.<n>We name our best-performing variant Dynamic Vision Aligner (DyVA)<n>We find DyVA to surpass both open-source and proprietary baselines, achieving state-of-the-art or comparable performance.
  arXiv  Detail & Related papers  (2025-10-01T13:07:05Z)
• From Generation to Generalization: Emergent Few-Shot Learning in Video Diffusion Models [65.0487600936788]
  Video Diffusion Models (VDMs) have emerged as powerful generative tools capable of synthesizing high-quality content.<n>We argue that VDMs naturally push to probe structured representations and an implicit understanding of the visual world.<n>Our method transforms each task into a visual transition, enabling the training of LoRA weights on short input- sequences.
  arXiv  Detail & Related papers  (2025-06-08T20:52:34Z)
• Object-level Scene Deocclusion [92.39886029550286]
  We present a new self-supervised PArallel visible-to-COmplete diffusion framework, named PACO, for object-level scene deocclusion. To train PACO, we create a large-scale dataset with 500k samples to enable self-supervised learning. Experiments on COCOA and various real-world scenes demonstrate the superior capability of PACO for scene deocclusion, surpassing the state of the arts by a large margin.
  arXiv  Detail & Related papers  (2024-06-11T20:34:10Z)
• Generative Camera Dolly: Extreme Monocular Dynamic Novel View Synthesis [43.02778060969546]
  We propose a controllable monocular dynamic view synthesis pipeline. Our model does not require depth as input, and does not explicitly model 3D scene geometry. We believe our framework can potentially unlock powerful applications in rich dynamic scene understanding, perception for robotics, and interactive 3D video viewing experiences for virtual reality.
  arXiv  Detail & Related papers  (2024-05-23T17:59:52Z)
• Dense Video Object Captioning from Disjoint Supervision [77.47084982558101]
  We propose a new task and model for dense video object captioning. This task unifies spatial and temporal localization in video. We show how our model improves upon a number of strong baselines for this new task.
  arXiv  Detail & Related papers  (2023-06-20T17:57:23Z)
• Learn the Force We Can: Enabling Sparse Motion Control in Multi-Object Video Generation [26.292052071093945]
  We propose an unsupervised method to generate videos from a single frame and a sparse motion input. Our trained model can generate unseen realistic object-to-object interactions. We show that YODA is on par with or better than state of the art video generation prior work in terms of both controllability and video quality.
  arXiv  Detail & Related papers  (2023-06-06T19:50:02Z)
• Pre-training Contextualized World Models with In-the-wild Videos for Reinforcement Learning [54.67880602409801]
  In this paper, we study the problem of pre-training world models with abundant in-the-wild videos for efficient learning of visual control tasks. We introduce Contextualized World Models (ContextWM) that explicitly separate context and dynamics modeling. Our experiments show that in-the-wild video pre-training equipped with ContextWM can significantly improve the sample efficiency of model-based reinforcement learning.
  arXiv  Detail & Related papers  (2023-05-29T14:29:12Z)
• Multi-Task Learning of Object State Changes from Uncurated Videos [55.60442251060871]
  We learn to temporally localize object state changes by observing people interacting with objects in long uncurated web videos. We show that our multi-task model achieves a relative improvement of 40% over the prior single-task methods. We also test our method on long egocentric videos of the EPIC-KITCHENS and the Ego4D datasets in a zero-shot setup.
  arXiv  Detail & Related papers  (2022-11-24T09:42:46Z)
• Self-supervised Amodal Video Object Segmentation [57.929357732733926]
  Amodal perception requires inferring the full shape of an object that is partially occluded. This paper develops a new framework of amodal Video object segmentation (SaVos)
  arXiv  Detail & Related papers  (2022-10-23T14:09:35Z)
• SAVi++: Towards End-to-End Object-Centric Learning from Real-World Videos [23.64091569954785]
  We introduce SAVi++, an object-centric video model which is trained to predict depth signals from a slot-based video representation. By using sparse depth signals obtained from LiDAR, SAVi++ is able to learn emergent object segmentation and tracking from videos in the real-world Open dataset.
  arXiv  Detail & Related papers  (2022-06-15T18:57:07Z)
• Support-Set Based Cross-Supervision for Video Grounding [98.29089558426399]
  Support-set Based Cross-Supervision (Sscs) module can improve existing methods during training phase without extra inference cost. The proposed Sscs module contains two main components, i.e., discriminative contrastive objective and generative caption objective. We extensively evaluate Sscs on three challenging datasets, and show that our method can improve current state-of-the-art methods by large margins.
  arXiv  Detail & Related papers  (2021-08-24T08:25:26Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.