Improving LLM Video Understanding with 16 Frames Per Second

• URL: http://arxiv.org/abs/2503.13956v1
• Date: Tue, 18 Mar 2025 06:48:08 GMT
• Title: Improving LLM Video Understanding with 16 Frames Per Second
• Authors: Yixuan Li, Changli Tang, Jimin Zhuang, Yudong Yang, Guangzhi Sun, Wei Li, Zejun Ma, Chao Zhang,
• Abstract summary: Existing methods rely on static features extracted from images sampled at a fixed low frame rate of frame-per-second (FPS) $leqslant$2, leading to critical visual information loss.<n>We introduce F-16, the first multimodal large language model (LLMs) designed for high-frame-rate video understanding.<n>F-16 efficiently captures dynamic visual features while preserving key semantic information.
• Score: 33.70837005629285
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Human vision is dynamic and continuous. However, in video understanding with multimodal large language models (LLMs), existing methods primarily rely on static features extracted from images sampled at a fixed low frame rate of frame-per-second (FPS) $\leqslant$2, leading to critical visual information loss. In this paper, we introduce F-16, the first multimodal LLM designed for high-frame-rate video understanding. By increasing the frame rate to 16 FPS and compressing visual tokens within each 1-second clip, F-16 efficiently captures dynamic visual features while preserving key semantic information. Experimental results demonstrate that higher frame rates considerably enhance video understanding across multiple benchmarks, providing a new approach to improving video LLMs beyond scaling model size or training data. F-16 achieves state-of-the-art performance among 7-billion-parameter video LLMs on both general and fine-grained video understanding benchmarks, such as Video-MME and TemporalBench. Furthermore, F-16 excels in complex spatiotemporal tasks, including high-speed sports analysis (\textit{e.g.}, basketball, football, gymnastics, and diving), outperforming SOTA proprietary visual models like GPT-4o and Gemini-1.5-pro. Additionally, we introduce a novel decoding method for F-16 that enables highly efficient low-frame-rate inference without requiring model retraining. Upon acceptance, we will release the source code, model checkpoints, and data.

Related papers

• An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes [85.00111442236499]
  This paper presents textbfQuicksviewer, an LMM with new perceiving paradigm that partitions a video of nontemporal density into varying cubes using Gumbel Softmax. We train the model from a language backbone through three progressive stages, each incorporating lengthy videos on average of 420s/1fps thanks to the perceiving efficiency. With only 0.8M total video-text samples for training, our model outperforms the direct baseline employing a fixed partitioning strategy by a maximum of 8.72 in accuracy.
  arXiv  Detail & Related papers  (2025-04-21T17:57:21Z)
• Token-Efficient Long Video Understanding for Multimodal LLMs [101.70681093383365]
  STORM is a novel architecture incorporating a dedicated temporal encoder between the image encoder and the Video-LLMs.<n>We show that STORM achieves state-of-the-art results across various long video understanding benchmarks.
  arXiv  Detail & Related papers  (2025-03-06T06:17:38Z)
• TS-LLaVA: Constructing Visual Tokens through Thumbnail-and-Sampling for Training-Free Video Large Language Models [52.590072198551944]
  Recent advances in multimodal Large Language Models (LLMs) have shown great success in understanding multi-modal contents. For video understanding tasks, training-based video LLMs are difficult to build due to the scarcity of high-quality, curated video-text paired data. In this work, we explore the limitations of the existing compression strategies for building a training-free video LLM.
  arXiv  Detail & Related papers  (2024-11-17T13:08:29Z)
• SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models [51.712700398020075]
  We propose a training-free video large language model (LLM) that can jointly capture detailed spatial semantics and long-range temporal context. This is realized by using a two-stream SlowFast design of inputs for Video LLMs to aggregate features from sampled frames in an effective way. Experimental results show that SF-LLaVA outperforms existing training-free methods on a wide range of video tasks.
  arXiv  Detail & Related papers  (2024-07-22T17:58:04Z)
• MiniGPT4-Video: Advancing Multimodal LLMs for Video Understanding with Interleaved Visual-Textual Tokens [36.02433030551474]
  MiniGPT4-Video is a multimodal Large Language Model (LLM) designed specifically for video understanding. MiniGPT4-video does not only consider visual content but also incorporates textual conversations, allowing the model to effectively answer queries involving both visual and text components.
  arXiv  Detail & Related papers  (2024-04-04T12:46:01Z)
• Video-LaVIT: Unified Video-Language Pre-training with Decoupled Visual-Motional Tokenization [52.63845811751936]
  Video pre-training is challenging due to the modeling of its dynamics video. In this paper, we address such limitations in video pre-training with an efficient video decomposition. Our framework is both capable of comprehending and generating image and video content, as demonstrated by its performance across 13 multimodal benchmarks.
  arXiv  Detail & Related papers  (2024-02-05T16:30:49Z)
• A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames [57.758863967770594]
  We build on the common paradigm of transferring large-scale, image--text models to video via shallow temporal fusion.<n>We expose two limitations to the approach: (1) decreased spatial capabilities, likely due to poor video--language alignment in standard video datasets, and (2) higher memory consumption, bottlenecking the number of frames that can be processed.
  arXiv  Detail & Related papers  (2023-12-12T16:10:19Z)
• Efficient Cross-Modal Video Retrieval with Meta-Optimized Frames [39.03408879727955]
  Cross-modal video retrieval aims to retrieve semantically relevant videos given a text as a query. A common and simple solution is to uniformly sample a small number of frames from the video as input to ViT. To get free from this trade-off, this paper introduces an automatic video compression method based on a bilevel optimization program.
  arXiv  Detail & Related papers  (2022-10-16T05:35:00Z)

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.