FALCONEye: Finding Answers and Localizing Content in ONE-hour-long videos with multi-modal LLMs

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Plou, Carlos, Borja, Cesar, Martinez-Cantin, Ruben, Murillo, Ana C.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908753695080448
author Plou, Carlos
Borja, Cesar
Martinez-Cantin, Ruben
Murillo, Ana C.
author_facet Plou, Carlos
Borja, Cesar
Martinez-Cantin, Ruben
Murillo, Ana C.
contents Finding information in hour-long videos is a challenging task even for top-performing Vision Language Models (VLMs), as encoding visual content quickly exceeds available context windows. To tackle this challenge, we present FALCONEye, a novel video agent based on a training-free, model-agnostic meta-architecture composed of a VLM and a Large Language Model (LLM). FALCONEye answers open-ended questions using an exploration-based search algorithm guided by calibrated confidence from the VLM's answers. We also introduce the FALCON-Bench benchmark, extending Question Answering problem to Video Answer Search-requiring models to return both the answer and its supporting temporal window for open-ended questions in hour-long videos. With just a 7B VLM and a lightweight LLM, FALCONEye outscores all open-source 7B VLMs and comparable agents in FALCON-Bench. It further demonstrates its generalization capability in MLVU benchmark with shorter videos and different tasks, surpassing GPT-4o on single-detail tasks while slashing inference cost by roughly an order of magnitude.
format Preprint
id arxiv_https___arxiv_org_abs_2503_19850
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FALCONEye: Finding Answers and Localizing Content in ONE-hour-long videos with multi-modal LLMs
Plou, Carlos
Borja, Cesar
Martinez-Cantin, Ruben
Murillo, Ana C.
Computer Vision and Pattern Recognition
Finding information in hour-long videos is a challenging task even for top-performing Vision Language Models (VLMs), as encoding visual content quickly exceeds available context windows. To tackle this challenge, we present FALCONEye, a novel video agent based on a training-free, model-agnostic meta-architecture composed of a VLM and a Large Language Model (LLM). FALCONEye answers open-ended questions using an exploration-based search algorithm guided by calibrated confidence from the VLM's answers. We also introduce the FALCON-Bench benchmark, extending Question Answering problem to Video Answer Search-requiring models to return both the answer and its supporting temporal window for open-ended questions in hour-long videos. With just a 7B VLM and a lightweight LLM, FALCONEye outscores all open-source 7B VLMs and comparable agents in FALCON-Bench. It further demonstrates its generalization capability in MLVU benchmark with shorter videos and different tasks, surpassing GPT-4o on single-detail tasks while slashing inference cost by roughly an order of magnitude.
title FALCONEye: Finding Answers and Localizing Content in ONE-hour-long videos with multi-modal LLMs
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.19850