Beyond Similarity: Foundation Models as an Efficient Backbone for Training-Free Composed Video Retrieval

• URL: http://arxiv.org/abs/2609.10008v1
• Date: Wed, 09 Sep 2026 10:38:25 GMT
• Title: Beyond Similarity: Foundation Models as an Efficient Backbone for Training-Free Composed Video Retrieval
• Abstract summary: Composed video retrieval (CoVR) searches a gallery for the target video that realizes a natural-language modification of a source clip.<n>We propose that frozen foundation models should instead occupy complementary roles, with inference depth adapted to query difficulty.<n>Our method reaches state-of-the-art performance among training-free approaches, with 89.55 and 93.43 R@1 on-WebVid-CoVR and CoVR-R, respectively.
• Score: 10.716619269996677
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Composed video retrieval (CoVR) searches a gallery for the target video that realizes a natural-language modification of a source clip. However, at gallery scale, this creates a fundamental tension: compact embeddings enable efficient, reusable search but can miss the transient actions, state changes, and subtle constraints that demand fine-grained video reasoning, whereas applying large multimodal models uniformly sacrifices scalability. To address these limitations, we propose that frozen foundation models should instead occupy complementary roles, with inference depth adapted to query difficulty. Based on this premise, we introduce \methodname{}, a framework for training-free \methodexpansion{}. Specifically, a composed-query embedding first searches reusable video-only gallery representations; uncertain queries undergo bounded reranking and candidate expansion; ambiguous edits trigger target-description generation; and only close leading candidates reach multimodal verification. To support these roles, frame selection, spatial resolution, and time cues are adapted to each stage. Across complete target-gallery evaluations, our method reaches state-of-the-art performance among training-free approaches, with 89.55 and 93.43 R@1 on Dense-WebVid-CoVR and CoVR-R, respectively (with more than +35\% and +25\% absolute margins to the closest counterpart). These results show that adaptively orchestrating foundation-model capabilities can combine scalable retrieval with fine-grained reasoning without task-specific training. The source code and all relevant guidelines are available on https://github.com/demidovd98/CoVRAGE.

Related papers

• A Paradigm Shift: Fully End-to-End Training for Temporal Sentence Grounding in Videos [27.879461637513984]
  Temporal sentence grounding in videos (TSGV) aims to localize a temporal segment that semantically corresponds to a sentence query from an untrimmed video.<n>Most current methods adopt pre-trained query-agnostic visual encoders for offline feature extraction, and the video backbones are frozen and not optimized for TSGV.<n>We propose a fully end-to-end paradigm that jointly optimize the video backbone and localization head.
  arXiv  Detail & Related papers  (2026-04-03T08:26:12Z)
• GMMFormer: Gaussian-Mixture-Model Based Transformer for Efficient Partially Relevant Video Retrieval [59.47258928867802]
  Given a text query, partially relevant video retrieval (PRVR) seeks to find videos containing pertinent moments in a database. This paper proposes GMMFormer, a Gaussian-Mixture-Model based Transformer which models clip representations implicitly. Experiments on three large-scale video datasets demonstrate the superiority and efficiency of GMMFormer.
  arXiv  Detail & Related papers  (2023-10-08T15:04:50Z)
• Transform-Equivariant Consistency Learning for Temporal Sentence Grounding [66.10949751429781]
  We introduce a novel Equivariant Consistency Regulation Learning framework to learn more discriminative representations for each video. Our motivation comes from that the temporal boundary of the query-guided activity should be consistently predicted. In particular, we devise a self-supervised consistency loss module to enhance the completeness and smoothness of the augmented video.
  arXiv  Detail & Related papers  (2023-05-06T19:29:28Z)
• DiffusionRet: Generative Text-Video Retrieval with Diffusion Model [56.03464169048182]
  Existing text-video retrieval solutions focus on maximizing the conditional likelihood, i.e., p(candidates|query) We creatively tackle this task from a generative viewpoint and model the correlation between the text and the video as their joint probability p(candidates,query) This is accomplished through a diffusion-based text-video retrieval framework (DiffusionRet), which models the retrieval task as a process of gradually generating joint distribution from noise.
  arXiv  Detail & Related papers  (2023-03-17T10:07:19Z)
• UATVR: Uncertainty-Adaptive Text-Video Retrieval [90.8952122146241]
  A common practice is to transfer text-video pairs to the same embedding space and craft cross-modal interactions with certain entities. We propose an Uncertainty-language Text-Video Retrieval approach, termed UATVR, which models each look-up as a distribution matching procedure.
  arXiv  Detail & Related papers  (2023-01-16T08:43:17Z)
• Deep Model Reassembly [60.6531819328247]
  We explore a novel knowledge-transfer task, termed as Deep Model Reassembly (DeRy) The goal of DeRy is to first dissect each model into distinctive building blocks, and then selectively reassemble the derived blocks to produce customized networks. We demonstrate that on ImageNet, the best reassemble model achieves 78.6% top-1 accuracy without fine-tuning.
  arXiv  Detail & Related papers  (2022-10-24T10:16:13Z)
• Learning Salient Boundary Feature for Anchor-free Temporal Action Localization [81.55295042558409]
  Temporal action localization is an important yet challenging task in video understanding. We propose the first purely anchor-free temporal localization method. Our model includes (i) an end-to-end trainable basic predictor, (ii) a saliency-based refinement module, and (iii) several consistency constraints.
  arXiv  Detail & Related papers  (2021-03-24T12:28:32Z)
• Frame-wise Cross-modal Matching for Video Moment Retrieval [32.68921139236391]
  Video moment retrieval targets at retrieving a moment in a video for a given language query. The challenges of this task include 1) the requirement of localizing the relevant moment in an untrimmed video, and 2) bridging the semantic gap between textual query and video contents. We propose an Attentive Cross-modal Relevance Matching model which predicts the temporal boundaries based on an interaction modeling.
  arXiv  Detail & Related papers  (2020-09-22T10:25:41Z)
• Perceptron Synthesis Network: Rethinking the Action Scale Variances in Videos [48.57686258913474]
  Video action recognition has been partially addressed by the CNNs stacking of fixed-size 3D kernels. We propose to learn the optimal-scale kernels from the data. An textitaction perceptron synthesizer is proposed to generate the kernels from a bag of fixed-size kernels.
  arXiv  Detail & Related papers  (2020-07-22T14:22:29Z)

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.