paper-with-me

홈 › Papers

Unifying Specialized Visual Encoders for Video Language Models

2025-01-02 · Jihoon Chung, Tyler Zhu, Max Gonzalez Saez-Diez, Juan Carlos Niebles, Honglu Zhou, Olga Russakovsky

The recent advent of Large Language Models (LLMs) has ushered sophisticated reasoning capabilities into the realm of video through Video Large Language Models (VideoLLMs). However, VideoLLMs currently rely on a single vision encoder for all of their visual processing, which limits the amount and type of visual information that can be conveyed to the LLM. Our method, MERV, Multi-Encoder Representation of Videos, instead leverages multiple frozen visual encoders to create a unified representation of a video, providing the VideoLLM with a comprehensive set of specialized visual knowledge. Spatio-temporally aligning the features from each encoder allows us to tackle a wider range of open-ended and multiple-choice video understanding questions and outperform prior state-of-the-art works. MERV is up to 3.7% better in accuracy than Video-LLaVA across the standard suite video understanding benchmarks, while also having a better Video-ChatGPT score. We also improve upon SeViLA, the previous best on zero-shot Perception Test accuracy, by 2.2%. MERV introduces minimal extra parameters and trains faster than equivalent single-encoder methods while parallelizing the visual processing. Finally, we provide qualitative evidence that MERV successfully captures domain knowledge from each of its encoders. Our results offer promising directions in utilizing multiple vision encoders for comprehensive video understanding.

📄 PDF Abstract BibTeX arXiv:2501.01426

Code (1)

princetonvisualai/merv 공식 구현 pytorch

Tasks

Multiple-choiceVideo Understanding

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

UniUGP: Unifying Understanding, Generation, and Planing For End-to-end Autonomous Driving

2025-12-10 · Hao Lu, Ziyang Liu, Guangfeng Jiang, Yuanfei Luo 외 arxiv

Autonomous driving (AD) systems struggle in long-tail scenarios due to limited world knowledge and weak visual dynamic modeling. Existing vision-language-action (VLA)-based methods cannot leverage unlabeled videos for vi…

Trajectory PlanningAutonomous DrivingVideo Generation

AnyGroundBench: A Specialized-Domain Benchmark for Video Grounding in Vision-Language Models

2026-07-02 · Rintaro Otsubo, Ryo Fujii, Reina Ishikawa, Taiki Kanaya 외 arxiv

Vision-Language Models (VLMs) have demonstrated immense promise in Spatio-Temporal Video Grounding (STVG). However, current evaluation protocols are largely confined to zero-shot assessments on general, daily-life benchm…

Spatio-Temporal Video GroundingZero-shot GeneralizationDomain Adaptation

AutoAD III: The Prequel -- Back to the Pixels

2024-04-22 · Tengda Han, Max Bain, Arsha Nagrani, Gül Varol 외

Generating Audio Description (AD) for movies is a challenging task that requires fine-grained visual understanding and an awareness of the characters and their names. Currently, visual language models for AD generation a…

AutoAD III: The Prequel - Back to the Pixels

2024-01-01 · CVPR 2024 1 · Tengda Han, Max Bain, Arsha Nagrani, Gül Varol 외

Generating Audio Description (AD) for movies is a challenging task that requires fine-grained visual understanding and an awareness of the characters and their names. Currently visual language models for AD generatio…

3D Aware Region Prompted Vision Language Model

2025-09-16 · An-Chieh Cheng, Yang Fu, Yukang Chen, Zhijian Liu 외 arxiv

We present Spatial Region 3D (SR-3D) aware vision-language model that connects single-view 2D images and multi-view 3D data through a shared visual token space. SR-3D supports flexible region prompting, allowing users to…

Scene UnderstandingSpatial Reasoning