VideoGEM: Training-free Action Grounding in Videos

• URL: http://arxiv.org/abs/2503.20348v1
• Date: Wed, 26 Mar 2025 09:20:30 GMT
• Title: VideoGEM: Training-free Action Grounding in Videos
• Authors: Felix Vogel, Walid Bousselham, Anna Kukleva, Nina Shvetsova, Hilde Kuehne,
• Abstract summary: We propose VideoGEM, the first training-free spatial action grounding method based on pretrained image- and video-language backbones.<n>We observe that high-level semantic concepts, such as actions, usually emerge in the higher layers of the image- and video-language models.<n>We introduce a prompt decomposition, processing action, verb, and object prompts separately, resulting in a better spatial localization of actions.
• Score: 20.477666113083682
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Vision-language foundation models have shown impressive capabilities across various zero-shot tasks, including training-free localization and grounding, primarily focusing on localizing objects in images. However, leveraging those capabilities to localize actions and events in videos is challenging, as actions have less physical outline and are usually described by higher-level concepts. In this work, we propose VideoGEM, the first training-free spatial action grounding method based on pretrained image- and video-language backbones. Namely, we adapt the self-self attention formulation of GEM to spatial activity grounding. We observe that high-level semantic concepts, such as actions, usually emerge in the higher layers of the image- and video-language models. We, therefore, propose a layer weighting in the self-attention path to prioritize higher layers. Additionally, we introduce a dynamic weighting method to automatically tune layer weights to capture each layer`s relevance to a specific prompt. Finally, we introduce a prompt decomposition, processing action, verb, and object prompts separately, resulting in a better spatial localization of actions. We evaluate the proposed approach on three image- and video-language backbones, CLIP, OpenCLIP, and ViCLIP, and on four video grounding datasets, V-HICO, DALY, YouCook-Interactions, and GroundingYouTube, showing that the proposed training-free approach is able to outperform current trained state-of-the-art approaches for spatial video grounding.

Related papers

• SMILE: Infusing Spatial and Motion Semantics in Masked Video Learning [50.98341607245458]
  Masked video modeling is an effective paradigm for video self-supervised learning (SSL) This paper introduces a novel SSL approach for video representation learning, dubbed as SMILE, by infusing both spatial and motion semantics. We establish a new self-supervised video learning paradigm capable of learning strong video representations without requiring any natural video data.
  arXiv  Detail & Related papers  (2025-04-01T08:20:55Z)
• ActPrompt: In-Domain Feature Adaptation via Action Cues for Video Temporal Grounding [40.60371529725805]
  We propose an efficient preliminary in-domain fine-tuning paradigm for feature adaptation. We introduce Action-Cue-Injected Temporal Prompt Learning (ActPrompt), which injects action cues into the image encoder of VLM for better discovering action-sensitive patterns.
  arXiv  Detail & Related papers  (2024-08-13T04:18:32Z)
• Puppet-Master: Scaling Interactive Video Generation as a Motion Prior for Part-Level Dynamics [67.97235923372035]
  We present Puppet-Master, an interactive video generative model that can serve as a motion prior for part-level dynamics. At test time, given a single image and a sparse set of motion trajectories, Puppet-Master can synthesize a video depicting realistic part-level motion faithful to the given drag interactions.
  arXiv  Detail & Related papers  (2024-08-08T17:59:38Z)
• Open-Vocabulary Spatio-Temporal Action Detection [59.91046192096296]
  Open-vocabulary-temporal action detection (OV-STAD) is an important fine-grained video understanding task. OV-STAD requires training a model on a limited set of base classes with box and label supervision. To better adapt the holistic VLM for the fine-grained action detection task, we carefully fine-tune it on the localized video region-text pairs.
  arXiv  Detail & Related papers  (2024-05-17T14:52:47Z)
• AffordanceLLM: Grounding Affordance from Vision Language Models [36.97072698640563]
  Affordance grounding refers to the task of finding the area of an object with which one can interact. Much of the knowledge is hidden and beyond the image content with the supervised labels from a limited training set. We make an attempt to improve the generalization capability of the current affordance grounding by taking the advantage of the rich world, abstract, and human-object-interaction knowledge.
  arXiv  Detail & Related papers  (2024-01-12T03:21:02Z)
• Language-based Action Concept Spaces Improve Video Self-Supervised Learning [8.746806973828738]
  We introduce language tied self-supervised learning to adapt an image CLIP model to the video domain. A backbone modified for temporal modeling is trained under self-distillation settings with train objectives operating in an action concept space. Our approach improves zero-shot and linear probing performance on three action recognition benchmarks.
  arXiv  Detail & Related papers  (2023-07-20T14:47:50Z)
• Dense Video Object Captioning from Disjoint Supervision [77.47084982558101]
  We propose a new task and model for dense video object captioning. This task unifies spatial and temporal localization in video. We show how our model improves upon a number of strong baselines for this new task.
  arXiv  Detail & Related papers  (2023-06-20T17:57:23Z)
• Part-level Action Parsing via a Pose-guided Coarse-to-Fine Framework [108.70949305791201]
  Part-level Action Parsing (PAP) aims to not only predict the video-level action but also recognize the frame-level fine-grained actions or interactions of body parts for each person in the video. In particular, our framework first predicts the video-level class of the input video, then localizes the body parts and predicts the part-level action. Our framework achieves state-of-the-art performance and outperforms existing methods over a 31.10% ROC score.
  arXiv  Detail & Related papers  (2022-03-09T01:30:57Z)
• Look at What I'm Doing: Self-Supervised Spatial Grounding of Narrations in Instructional Videos [78.34818195786846]
  We introduce the task of spatially localizing narrated interactions in videos. Key to our approach is the ability to learn to spatially localize interactions with self-supervision on a large corpus of videos with accompanying transcribed narrations. We propose a multilayer cross-modal attention network that enables effective optimization of a contrastive loss during training.
  arXiv  Detail & Related papers  (2021-10-20T14:45:13Z)

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.