arXiv Computer Vision October 19, 2026 By Ruizhi Chen, Zining Fan, Yudi Guo, Jiang Liu, Zhiyu Lin, Zhongjiang He, Shouqian Shi, Hao Sun

VTrajEmbed: Aligning Text with Video Object Trajectories via Spatiotemporal Embeddings

Read the original on arXiv Computer Vision →

VTrajEmbed is an embedding model that aligns text with video object trajectories by decomposing videos into individual trajectories and embedding each trajectory separately. It introduces a Slow‑Fast Trajectory Encoding dual‑branch architecture that captures high‑frequency motion and scene context, and a Mask‑Guided Contrastive Learning objective that prevents false negatives during training. The model is evaluated on a new trajectory‑level benchmark, Text2Traj‑Bench, and outperforms existing large‑language‑model embeddings by significant margins on both text‑to‑trajectory and trajectory‑to‑text retrieval tasks.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

arXiv Machine Learning

Jul 2

MoVA: Learning Asymmetric Dual Projections for Modular Long Video-Text Alignment

arXiv:2607. 00858v1 Announce Type: cross Abstract: Contrastive pre-training has propelled video-text alignment, yet models often inherit the critical limitations of their image-text predecessors like CLIP, resulting in entangled representations.

By Peiyuan Zhu, Shaoan Xie, Zijian Li, Yifan Shen, Namrata Deka, Harsh Shrivastava, Guangyi Chen, Kun Zhang

multimodal safety

arXiv Machine Learning

Aug 14

TraVEL: Trajectory-Guided Video Embedding Learning for Driving-Video Retrieval

arXiv:2608. 13495v1 Announce Type: cross Abstract: Efficiently retrieving relevant clips from large-scale driving logs is essential for data curation, model development, and safety analysis.

By Yi-Chung Chen, Philip Jacobson, Tom Lampo, Yiren Lu, Jin Yao, David I. Inouye, Jing Gao, Danhua Guo, Burhan Yaman

rag fine-tuning multimodal benchmarks safety

arXiv AI

Jun 30

UniMotion: A Unified Framework for Motion-Text-Vision Understanding and Generation

arXiv:2603. 22282v2 Announce Type: replace-cross Abstract: We present UniMotion, to our knowledge the first unified framework for simultaneous understanding and generation of human motion, natural language, and RGB images within a single architecture.

By Ziyi Wang, Xinshun Wang, Shuang Chen, Yang Cong, Mengyuan Liu

llms nlp efficiency multimodal benchmarks safety

arXiv AI

Oct 16

Where, How, and When: Query-Guided Grounding for Video Understanding

The paper introduces query‑guided grounding methods for video understanding, covering object‑level, relationship‑level, and moment‑level tasks. For object grounding, QDETRv and TOViT enable one‑shot and few‑shot localization of unseen objects using query‑context fusion, memory modules, and language‑aligned vision encoders. Relationship grounding is addressed with a few‑shot referring framework that optimizes a T‑partite random field, while moment grounding is tackled by the Moment Alignment Transformer (MATR), which employs dynamic time warping and self‑supervised pretraining for video‑to‑video and text‑to‑video retrieval.

By Yogesh Kumar

llms rag diffusion computer-vision safety

arXiv Computer Vision

Sep 3

TAME: Temporal-Aware Mixture-of-Experts for Text-Video Retrieval

TAME is a CLIP‑based framework for Text‑Video Retrieval that incorporates temporal modeling through three key innovations: sparse Mixture‑of‑Experts layers with frame‑consistent routing, Frame‑Temporal tokens that aggregate cross‑frame information, and a Cross‑Temporal Interaction and Aggregation module for refining frame‑wise similarities. These components enable the model to capture both local visual patterns and long‑range temporal dependencies, leading to consistent performance gains over CLIP‑based baselines on multiple TVR benchmarks, including a 4.0 R@1 improvement on MSR‑VTT. The code is publicly available on GitHub.

By Uicheol Jung, Juyoung Hong, Hojung Kwon, Yukyung Choi

multimodal benchmarks safety

arXiv Computer Vision

Aug 31

Locate Anything in Videos: Rethinking Efficient Generative Spatio-Temporal Video Grounding

The paper introduces Parallel Tube Decoding (PTD), a generative approach for spatio‑temporal video grounding that splits the task into a temporal block and simultaneous time‑conditioned spatial blocks, eliminating token‑level and trajectory‑level dependencies. PTD uses Decoupled Block Attention to allow parallel spatial generation while maintaining shared video‑query context, and incorporates localization‑aware policy optimization for temporal boundaries and spatial geometry. Experiments on VidSTG show PTD cuts tube completion latency by 79× and boosts spatial decoding throughput by 92× compared to autoregressive decoding, while improving grounding accuracy and performing well on related tasks such as temporal grounding, VideoQA, and referring video object tracking.

By Hanoona Rasheed, Haania Siddiqui, Ming-Hsuan Yang, Fahad Shahbaz Khan, Salman Khan

llms efficiency multimodal