TrajTok: Learning Trajectory Tokens enables better Video Understanding

• URL: http://arxiv.org/abs/2602.22779v1
• Date: Thu, 26 Feb 2026 09:15:34 GMT
• Title: TrajTok: Learning Trajectory Tokens enables better Video Understanding
• Authors: Chenhao Zheng, Jieyu Zhang, Jianing Zhang, Weikai Huang, Ashutosh Kumar, Quan Kong, Oncel Tuzel, Chun-Liang Li, Ranjay Krishna,
• Abstract summary: Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens.<n>We propose TrajTok, an end-to-end video tokenizer module that is fully integrated and co-trained with video models for a downstream objective.<n>We show that it can be seamlessly integrated as either a probing head for pretrained visual features (TrajAdapter) or an alignment connector in vision-language models (TrajVLM) with especially strong performance in long-video reasoning.
• Score: 63.1260672430712
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens. This severely limits video efficiency and scalability. While recent trajectory-based tokenizers offer a promising solution by decoupling video duration from token count, they rely on complex external segmentation and tracking pipelines that are slow and task-agnostic. We propose TrajTok, an end-to-end video tokenizer module that is fully integrated and co-trained with video models for a downstream objective, dynamically adapting its token granularity to semantic complexity, independent of video duration. TrajTok contains a unified segmenter that performs implicit clustering over pixels in both space and time to directly produce object trajectories in a single forward pass. By prioritizing downstream adaptability over pixel-perfect segmentation fidelity, TrajTok is lightweight and efficient, yet empirically improves video understanding performance. With TrajTok, we implement a video CLIP model trained from scratch (TrajViT2). It achieves the best accuracy at scale across both classification and retrieval benchmarks, while maintaining efficiency comparable to the best token-merging methods. TrajTok also proves to be a versatile component beyond its role as a tokenizer. We show that it can be seamlessly integrated as either a probing head for pretrained visual features (TrajAdapter) or an alignment connector in vision-language models (TrajVLM) with especially strong performance in long-video reasoning.

Related papers

• Fast SAM2 with Text-Driven Token Pruning [52.8350457627401]
  Segment Anything Model 2 (SAM2), a vision computation model has significantly advanced in prompt-driven video object segmentation.<n>SAM2 pipelines propagate all visual tokens produced by the image encoder through downstream temporal reasoning modules, regardless of their relevance to the target object.<n>We introduce a text-guided token pruning framework that improves inference efficiency by selectively reducing token density prior to temporal propagation.
  arXiv  Detail & Related papers  (2025-12-24T18:59:05Z)
• FLoC: Facility Location-Based Efficient Visual Token Compression for Long Video Understanding [55.700832127331324]
  FLoC is an efficient visual token compression framework based on the facility location function.<n>Our method achieves remarkable efficiency gains by swiftly selecting a compact subset of tokens.<n>Our approach is training-free, model-agnostic, and query-agnostic, providing a versatile solution.
  arXiv  Detail & Related papers  (2025-10-31T17:29:39Z)
• One Trajectory, One Token: Grounded Video Tokenization via Panoptic Sub-object Trajectory [25.726492556054904]
  We introduce grounded video tokenization, a paradigm that organizes tokens based on panoptic sub-object trajectories rather than fixed patches.<n>We propose TrajViT, a video encoder that extracts object trajectories and converts them into semantically meaningful tokens.<n>We show TrajViT as a stronger model than ViT3D for being the video encoder for modern VideoLLM.
  arXiv  Detail & Related papers  (2025-05-29T16:25:35Z)
• VQToken: Neural Discrete Token Representation Learning for Extreme Token Reduction in Video Large Language Models [35.38573641029626]
  We introduce the novel task of Extreme Short Token Reduction, which aims to represent entire videos using a minimal set of discrete tokens.<n>On the Extreme Short Token Reduction task, our VQToken compresses sequences to just 0.07 percent of their original length while incurring only a 0.66 percent drop in accuracy on the NextQA-MC benchmark.
  arXiv  Detail & Related papers  (2025-03-21T09:46:31Z)
• Learning Motion and Temporal Cues for Unsupervised Video Object Segmentation [49.113131249753714]
  We propose an efficient algorithm, termed MTNet, which concurrently exploits motion and temporal cues.<n> MTNet is devised by effectively merging appearance and motion features during the feature extraction process within encoders.<n>We employ a cascade of decoders all feature levels across all feature levels to optimally exploit the derived features.
  arXiv  Detail & Related papers  (2025-01-14T03:15:46Z)
• Beyond Training: Dynamic Token Merging for Zero-Shot Video Understanding [11.211803499867639]
  We propose DYTO, a novel dynamic token merging framework for zero-shot video understanding.<n> DYTO integrates a hierarchical frame selection and a bipartite token merging strategy to dynamically cluster key frames and selectively compress token sequences.<n>Experiments demonstrate the effectiveness of DYTO, achieving superior performance compared to both fine-tuned and training-free methods.
  arXiv  Detail & Related papers  (2024-11-21T18:30:11Z)
• CenterCLIP: Token Clustering for Efficient Text-Video Retrieval [67.21528544724546]
  In CLIP, the essential visual tokenization process, which produces discrete visual token sequences, generates many homogeneous tokens due to the redundancy nature of consecutive frames in videos. This significantly increases computation costs and hinders the deployment of video retrieval models in web applications. In this paper, we design a multi-segment token clustering algorithm to find the most representative tokens and drop the non-essential ones.
  arXiv  Detail & Related papers  (2022-05-02T12:02:09Z)
• CompFeat: Comprehensive Feature Aggregation for Video Instance Segmentation [67.17625278621134]
  Video instance segmentation is a complex task in which we need to detect, segment, and track each object for any given video. Previous approaches only utilize single-frame features for the detection, segmentation, and tracking of objects. We propose a novel comprehensive feature aggregation approach (CompFeat) to refine features at both frame-level and object-level with temporal and spatial context information.
  arXiv  Detail & Related papers  (2020-12-07T00:31:42Z)
• Fast Video Object Segmentation With Temporal Aggregation Network and Dynamic Template Matching [67.02962970820505]
  We introduce "tracking-by-detection" into Video Object (VOS) We propose a new temporal aggregation network and a novel dynamic time-evolving template matching mechanism to achieve significantly improved performance. We achieve new state-of-the-art performance on the DAVIS benchmark without complicated bells and whistles in both speed and accuracy, with a speed of 0.14 second per frame and J&F measure of 75.9% respectively.
  arXiv  Detail & Related papers  (2020-07-11T05:44:16Z)

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.