Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder

• URL: http://arxiv.org/abs/2506.22880v1
• Date: Sat, 28 Jun 2025 13:30:36 GMT
• Title: Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder
• Authors: Dang Jisheng, Wu Xudong, Wang Bimei, Lv Ning, Chen Jiayu, Jingwen Zhao, Yichu liu, Jizhao Liu, Juncheng Li, Teng Wang,
• Abstract summary: Video segmenter and grounder approaches, exemplified by Sa2VA, directly fuse features within segmentation models.<n>This often results in an undesirable entanglement of dynamic visual information and static semantics, thereby degrading segmentation accuracy.<n>We propose DeSa2VA, a decoupling-enhanced prompting scheme integrating text pre-training and a linear decoupling module to address the information processing limitations inherent in SAM-2.
• Score: 5.57393627015653
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing video segmenter and grounder approaches, exemplified by Sa2VA, directly fuse features within segmentation models. This often results in an undesirable entanglement of dynamic visual information and static semantics, thereby degrading segmentation accuracy. To systematically mitigate this issue, we propose DeSa2VA, a decoupling-enhanced prompting scheme integrating text pre-training and a linear decoupling module to address the information processing limitations inherent in SAM-2. Specifically, first, we devise a pre-training paradigm that converts textual ground-truth labels into point-level prompts while generating corresponding text masks. These masks are refined through a hybrid loss function to strengthen the model's semantic grounding capabilities. Next, we employ linear projection to disentangle hidden states that generated by a large language model into distinct textual and visual feature subspaces. Finally, a dynamic mask fusion strategy synergistically combines these decoupled features through triple supervision from predicted text/visual masks and ground-truth annotations. Extensive experiments demonstrate state-of-the-art performance across diverse tasks, including image segmentation, image question answering, video segmentation, and video question answering. Our codes are available at https://github.com/longmalongma/DeSa2VA.

Related papers

• TASeg: Text-aware RGB-T Semantic Segmentation based on Fine-tuning Vision Foundation Models [26.983562312613877]
  We propose a text-aware RGB-T segmentation framework by using Low-Rank Adaptation (LoRA) fine-tuning technology to adapt vision foundation models.<n>Specifically, we propose a Dynamic Feature Fusion Module (DFFM) in the image encoder, which effectively merges features from multiple visual modalities while freezing SAM's original transformer blocks.
  arXiv  Detail & Related papers  (2025-06-27T07:34:28Z)
• LlamaSeg: Image Segmentation via Autoregressive Mask Generation [46.17509085054758]
  We present LlamaSeg, a visual autoregressive framework that unifies multiple image segmentation tasks via natural language instructions.<n>We reformulate image segmentation as a visual generation problem, representing masks as "visual" tokens and employing a LLaMA-style Transformer to predict them directly from image inputs.
  arXiv  Detail & Related papers  (2025-05-26T02:22:41Z)
• Mask$^2$DiT: Dual Mask-based Diffusion Transformer for Multi-Scene Long Video Generation [62.56037816595509]
  Mask$2$DiT establishes fine-grained, one-to-one alignment between video segments and their corresponding text annotations.<n>This attention mechanism enables precise segment-level textual-to-visual alignment.<n>Mask$2$DiT excels in maintaining visual consistency across segments while ensuring semantic alignment between each segment and its corresponding text description.
  arXiv  Detail & Related papers  (2025-03-25T17:46:50Z)
• InvSeg: Test-Time Prompt Inversion for Semantic Segmentation [33.60580908728705]
  InvSeg is a test-time prompt inversion method that tackles open-vocabulary semantic segmentation.<n>We introduce Contrastive Soft Clustering (CSC) to align derived masks with the image's structure information.<n>InvSeg learns context-rich text prompts in embedding space and achieves accurate semantic alignment across modalities.
  arXiv  Detail & Related papers  (2024-10-15T10:20:31Z)
• OneRef: Unified One-tower Expression Grounding and Segmentation with Mask Referring Modeling [80.85164509232261]
  We propose OneRef, a minimalist referring framework built on the modality-shared one-tower transformer. To modeling the referential relationship, we introduce a novel MVLM paradigm called Mask Referring Modeling (MRefM) Within MRefM, we propose a referring-aware dynamic image masking strategy that is aware of the referred region.
  arXiv  Detail & Related papers  (2024-10-10T15:18:19Z)
• SegVG: Transferring Object Bounding Box to Segmentation for Visual Grounding [56.079013202051094]
  We present SegVG, a novel method transfers the box-level annotation as signals to provide an additional pixel-level supervision for Visual Grounding. This approach allows us to iteratively exploit the annotation as signals for both box-level regression and pixel-level segmentation.
  arXiv  Detail & Related papers  (2024-07-03T15:30:45Z)
• CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor [18.288738950822342]
  Mask labels are labor-intensive, which limits the number of categories in segmentation datasets. We introduce a novel recurrent framework that progressively filters out irrelevant texts and enhances mask quality without training efforts. Experiments show that our method outperforms not only the training-free counterparts, but also those fine-tuned with millions of data samples.
  arXiv  Detail & Related papers  (2023-12-12T19:00:04Z)
• Video-Text Pre-training with Learned Regions [59.30893505895156]
  Video-Text pre-training aims at learning transferable representations from large-scale video-text pairs. We propose a module for videotext-learning, RegionLearner, which can take into account the structure of objects during pre-training on large-scale video-text pairs.
  arXiv  Detail & Related papers  (2021-12-02T13:06:53Z)
• VIMPAC: Video Pre-Training via Masked Token Prediction and Contrastive Learning [82.09856883441044]
  Video understanding relies on perceiving the global content modeling its internal connections. We propose a block-wise strategy where we mask neighboring video tokens in both spatial and temporal domains. We also add an augmentation-free contrastive learning method to further capture global content.
  arXiv  Detail & Related papers  (2021-06-21T16:48:19Z)
• Segmenter: Transformer for Semantic Segmentation [79.9887988699159]
  We introduce Segmenter, a transformer model for semantic segmentation. We build on the recent Vision Transformer (ViT) and extend it to semantic segmentation. It outperforms the state of the art on the challenging ADE20K dataset and performs on-par on Pascal Context and Cityscapes.
  arXiv  Detail & Related papers  (2021-05-12T13:01:44Z)

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.