PEEK: Picking Essential frames via Efficient Knowledge distillation

• URL: http://arxiv.org/abs/2605.31029v1
• Date: Fri, 29 May 2026 09:01:56 GMT
• Title: PEEK: Picking Essential frames via Efficient Knowledge distillation
• Abstract summary: Adaptive frame sampling has emerged as a promising approach for selecting the most informative frames from a video.<n>We introduce PEEK, an efficient dynamic frame sampling method that distills caption-conditioned frame relevance rankings from a stronger teacher model.<n>We find that, overall, on ActivityNet Captions and MSR-VTT, our method outperforms state-of-the-art methods.
• Score: 2.0117085784914623
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Video-language models can process only a limited number of frames, making frame selection a key bottleneck for efficient video captioning. Most captioning pipelines still rely on uniform sampling, which is computationally cheap but agnostic to visual content. Adaptive frame sampling has recently emerged as a promising approach for selecting the most informative frames from a video; however, existing methods remain computationally expensive. We introduce PEEK, an efficient dynamic frame sampling method that distills caption-conditioned frame relevance rankings from a stronger teacher model into a lightweight temporal model that operates only on visual content. We find that, overall, on ActivityNet Captions and MSR-VTT, our method outperforms state-of-the-art methods across all evaluated downstream vision language models, especially when only one or two frames are selected for captioning, obtaining the best CIDEr for most frame budgets. On ActivityNet Captions, PEEK is particularly strong, winning 14 out of 16 configurations. Zero-shot evaluation on MSR-VTT shows that our model transfers best at low frame budgets, while results at four and eight frames are more mixed as temporal coverage and visual diversity become increasingly competitive. Compared with recent adaptive baselines, PEEK is both more accurate in the low-budget regime and more efficient: it adds only $5.2\%$ to the captioning time, compared with $65.4\%$ for CSTA and $211.9\%$ for MaxInfo. We release our code and pre-trained checkpoint at https://github.com/momentslab/peek.

Related papers

• Coverage-Driven Adaptive Keyframe Selection for Video Understanding [12.078063042214573]
  We propose CSES, a training-free semantic selector that adaptively determines the numbers of frames to score and select.<n> CSES estimates the prominence of the frame-query to guide active acquisition and adapt the temporal coverage of each input.<n>It then formulates selection relevance as a coverage problem that jointly accounts for semantic relevance, temporal redundancy, and visual redundancy.
  arXiv  Detail & Related papers  (2026-08-01T15:24:06Z)
• GIFT: Global Irreplaceability Frame Targeting for Efficient Video Understanding [63.62445065977448]
  GIFT: Global Irreplaceability Frame Targeting is a training-free framework that selects frames by assessing their intrinsic irreplaceability.<n>We show that GIFT achieves a maximum average improvement of 12.5% across long-form video benchmarks on LLaVA-Video-7B compared to uniform sampling.
  arXiv  Detail & Related papers  (2026-03-26T06:21:41Z)
• FOCUS: Efficient Keyframe Selection for Long Video Understanding [26.44459739499484]
  Multimodal large language models (LMLMs) represent images and video frames as visual tokens.<n> FOCUS, Frame-Optimistic Confidence Upperbound Selection, is a model-agnostic selection module that selects frames under a strict token budget.<n>For videos longer than 20 minutes, it achieves an 11.9% gain in accuracy on LongVideoBenching benchmarks.
  arXiv  Detail & Related papers  (2025-10-31T08:41:13Z)
• Less is More: Token-Efficient Video-QA via Adaptive Frame-Pruning and Semantic Graph Integration [24.337139909108117]
  "Less is more" phenomenon where excessive frames can paradoxically degrade performance due to context dilution.<n>"Visual echoes" yield significant temporal redundancy, which we term 'visual echoes'<n>"AFP" employs an adaptive hierarchical clustering algorithm on a fused ResNet-50 and CLIP feature space to identify and merge these echoes into single representatives.<n>Our full approach demonstrates a drastic reduction in required frames by up to 86.9% and total input tokens by up to 83.2%.
  arXiv  Detail & Related papers  (2025-08-05T11:31:55Z)
• BOLT: Boost Large Vision-Language Model Without Training for Long-form Video Understanding [51.49345400300556]
  Large video-language models (VLMs) have demonstrated promising progress in various video understanding tasks.<n>Traditional approaches, such as uniform frame sampling, often inevitably allocate resources to irrelevant content.<n>We introduce BOLT, a method to BOost Large VLMs without additional Training through a comprehensive study of frame selection strategies.
  arXiv  Detail & Related papers  (2025-03-27T13:18:40Z)
• An Empirical Study of Frame Selection for Text-to-Video Retrieval [62.28080029331507]
  Text-to-video retrieval (TVR) aims to find the most relevant video in a large video gallery given a query text. Existing methods typically select a subset of frames within a video to represent the video content for TVR. In this paper, we make the first empirical study of frame selection for TVR.
  arXiv  Detail & Related papers  (2023-11-01T05:03:48Z)
• 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)
• Contrastive Video-Language Learning with Fine-grained Frame Sampling [54.542962813921214]
  FineCo is an approach to better learn video and language representations with a fine-grained contrastive objective operating on video frames. It helps distil a video by selecting the frames that are semantically equivalent to the text, improving cross-modal correspondence.
  arXiv  Detail & Related papers  (2022-10-10T22:48:08Z)
• Revealing Single Frame Bias for Video-and-Language Learning [115.01000652123882]
  We show that a single-frame trained model can achieve better performance than existing methods that use multiple frames for training. This result reveals the existence of a strong "static appearance bias" in popular video-and-language datasets. We propose two new retrieval tasks based on existing fine-grained action recognition datasets that encourage temporal modeling.
  arXiv  Detail & Related papers  (2022-06-07T16:28:30Z)
• OCSampler: Compressing Videos to One Clip with Single-step Sampling [82.0417131211353]
  We propose a framework named OCSampler to explore a compact yet effective video representation with one short clip. Our basic motivation is that the efficient video recognition task lies in processing a whole sequence at once rather than picking up frames sequentially.
  arXiv  Detail & Related papers  (2022-01-12T09:50:38Z)

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.