Unified Video Dense Prediction from Disjoint Data

• URL: http://arxiv.org/abs/2607.21592v1
• Date: Thu, 23 Jul 2026 17:59:50 GMT
• Title: Unified Video Dense Prediction from Disjoint Data
• Abstract summary: We introduce UniD, a unified video model that jointly predicts eight dense scene properties-depth, surface normals, semantic segmentation, boundaries, human parts, albedo, shading, and materials.<n>We propose a simple yet effective distillation step in which per-task experts supervise a unified backbone through lightweight task projectors.<n>UniD achieves competitive performance against per-task specialists and multi-task baselines.
• Score: 71.9957801898161
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Scene understanding requires simultaneous prediction about geometry, appearance, and semantics. However, existing task-specific annotations are fragmented across incompatible, domain-specific datasets. Current unified systems circumvent this by restricting training to fully co-annotated data, or by incurring the large computational cost of pseudo-labeling. To mitigate this, we introduce UniD, a unified video model that jointly predicts eight dense scene properties-depth, surface normals, semantic segmentation, boundaries, human parts, albedo, shading, and materials-all learned from disjoint, domain-specific datasets. We propose a simple yet effective distillation step in which per-task experts supervise a unified backbone through lightweight task projectors, eliminating the need for annotation overlap or pseudo-labeling. Our key insight is that the strong visual priors of a pretrained diffusion model are sufficient to bridge the domain gaps introduced by disjoint training sources, enabling robust generalization to scene-task combinations never seen during training. UniD achieves competitive performance against per-task specialists and multi-task baselines, with strong generalization to out-of-distribution scenarios and enhanced temporal and cross-task consistency. Code and video results are available at https://unid-video.github.io/.

Related papers

• Multi-Task Learning for Heterogeneous Prediction from Video Game State with Transfer Learning [41.132061576291065]
  Multi-task learning is a promising approach for prediction tasks derived from video game state data.<n>We study whether a shared model trained jointly across tasks in team-based multiplayer games can improve generalization.
  arXiv  Detail & Related papers  (2026-07-23T13:12:51Z)
• Tracking and Segmenting Anything in Any Modality [75.32774085793498]
  We propose a universal tracking and segmentation framework named SATA, which unifies a broad spectrum of tracking and segmentation subtasks with any modality input.<n> SATA demonstrates superior performance on 18 challenging tracking and segmentation benchmarks, offering a novel perspective for more generalizable video understanding.
  arXiv  Detail & Related papers  (2025-11-22T09:09:22Z)
• SpecDM: Hyperspectral Dataset Synthesis with Pixel-level Semantic Annotations [27.391859339238906]
  In this paper, we explore the potential of generative diffusion model in synthesizing hyperspectral images with pixel-level annotations.<n>To the best of our knowledge, it is the first work to generate high-dimensional HSIs with annotations.<n>We select two of the most widely used dense prediction tasks: semantic segmentation and change detection, and generate datasets suitable for these tasks.
  arXiv  Detail & Related papers  (2025-02-24T11:13:37Z)
• Comprehensive Generative Replay for Task-Incremental Segmentation with Concurrent Appearance and Semantic Forgetting [49.87694319431288]
  Generalist segmentation models are increasingly favored for diverse tasks involving various objects from different image sources. We propose a Comprehensive Generative (CGR) framework that restores appearance and semantic knowledge by synthesizing image-mask pairs. Experiments on incremental tasks (cardiac, fundus and prostate segmentation) show its clear advantage for alleviating concurrent appearance and semantic forgetting.
  arXiv  Detail & Related papers  (2024-06-28T10:05:58Z)
• Distribution Matching for Multi-Task Learning of Classification Tasks: a Large-Scale Study on Faces & Beyond [62.406687088097605]
  Multi-Task Learning (MTL) is a framework, where multiple related tasks are learned jointly and benefit from a shared representation space. We show that MTL can be successful with classification tasks with little, or non-overlapping annotations. We propose a novel approach, where knowledge exchange is enabled between the tasks via distribution matching.
  arXiv  Detail & Related papers  (2024-01-02T14:18:11Z)
• ULTRA-DP: Unifying Graph Pre-training with Multi-task Graph Dual Prompt [67.8934749027315]
  We propose a unified framework for graph hybrid pre-training which injects the task identification and position identification into GNNs. We also propose a novel pre-training paradigm based on a group of $k$-nearest neighbors.
  arXiv  Detail & Related papers  (2023-10-23T12:11:13Z)
• 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)
• Taskology: Utilizing Task Relations at Scale [28.09712466727001]
  We show that we can leverage the inherent relationships among collections of tasks, as they are trained jointly. explicitly utilizing the relationships between tasks allows improving their performance while dramatically reducing the need for labeled data. We demonstrate our framework on subsets of the following collection of tasks: depth and normal prediction, semantic segmentation, 3D motion and ego-motion estimation, and object tracking and 3D detection in point clouds.
  arXiv  Detail & Related papers  (2020-05-14T22:53:46Z)

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.