GateFusion: Hierarchical Gated Cross-Modal Fusion for Active Speaker Detection

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Wang, Yu, Ha, Juhyung, Ramirez, Frangil M., Wang, Yuchen, Crandall, David J.
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908718761771008
author Wang, Yu
Ha, Juhyung
Ramirez, Frangil M.
Wang, Yuchen
Crandall, David J.
author_facet Wang, Yu
Ha, Juhyung
Ramirez, Frangil M.
Wang, Yuchen
Crandall, David J.
contents Active Speaker Detection (ASD) aims to identify who is currently speaking in each frame of a video. Most state-of-the-art approaches rely on late fusion to combine visual and audio features, but late fusion often fails to capture fine-grained cross-modal interactions, which can be critical for robust performance in unconstrained scenarios. In this paper, we introduce GateFusion, a novel architecture that combines strong pretrained unimodal encoders with a Hierarchical Gated Fusion Decoder (HiGate). HiGate enables progressive, multi-depth fusion by adaptively injecting contextual features from one modality into the other at multiple layers of the Transformer backbone, guided by learnable, bimodally-conditioned gates. To further strengthen multimodal learning, we propose two auxiliary objectives: Masked Alignment Loss (MAL) to align unimodal outputs with multimodal predictions, and Over-Positive Penalty (OPP) to suppress spurious video-only activations. GateFusion establishes new state-of-the-art results on several challenging ASD benchmarks, achieving 77.8% mAP (+9.4%), 86.1% mAP (+2.9%), and 96.1% mAP (+0.5%) on Ego4D-ASD, UniTalk, and WASD benchmarks, respectively, and delivering competitive performance on AVA-ActiveSpeaker. Out-of-domain experiments demonstrate the generalization of our model, while comprehensive ablations show the complementary benefits of each component.
format Preprint
id arxiv_https___arxiv_org_abs_2512_15707
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle GateFusion: Hierarchical Gated Cross-Modal Fusion for Active Speaker Detection
Wang, Yu
Ha, Juhyung
Ramirez, Frangil M.
Wang, Yuchen
Crandall, David J.
Computer Vision and Pattern Recognition
Active Speaker Detection (ASD) aims to identify who is currently speaking in each frame of a video. Most state-of-the-art approaches rely on late fusion to combine visual and audio features, but late fusion often fails to capture fine-grained cross-modal interactions, which can be critical for robust performance in unconstrained scenarios. In this paper, we introduce GateFusion, a novel architecture that combines strong pretrained unimodal encoders with a Hierarchical Gated Fusion Decoder (HiGate). HiGate enables progressive, multi-depth fusion by adaptively injecting contextual features from one modality into the other at multiple layers of the Transformer backbone, guided by learnable, bimodally-conditioned gates. To further strengthen multimodal learning, we propose two auxiliary objectives: Masked Alignment Loss (MAL) to align unimodal outputs with multimodal predictions, and Over-Positive Penalty (OPP) to suppress spurious video-only activations. GateFusion establishes new state-of-the-art results on several challenging ASD benchmarks, achieving 77.8% mAP (+9.4%), 86.1% mAP (+2.9%), and 96.1% mAP (+0.5%) on Ego4D-ASD, UniTalk, and WASD benchmarks, respectively, and delivering competitive performance on AVA-ActiveSpeaker. Out-of-domain experiments demonstrate the generalization of our model, while comprehensive ablations show the complementary benefits of each component.
title GateFusion: Hierarchical Gated Cross-Modal Fusion for Active Speaker Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.15707