CORTEXA
← Browse
arxivcs.CVcs.SD2026-07-03

$C^3$ASD: Multi-Level Consistency-Driven Representation Learning

Jin Hong, Jisoo Park, Junseok Kwon

Active Speaker Detection determines whether a visible person in a video is speaking at each moment. While recent audio-visual fusion methods perform well on clean data, they degrade under real-world corruptions such as background noise, occlusion, or simultaneous modality degradation. We attribute this limitation to the absence of explicit consistency constraints that promote robust, semantically aligned representations across modalities. Without such guidance, models tend to learn fragile modality-specific shortcuts that fail under corrupted conditions. We propose $C^3$ASD, a multi-level consistency-driven framework with three complementary constraints: embedding-level inter-modality consistency aligns audio-visual representations during speech; sequence-level intra-modality consistency separates speaking and non-speaking clusters via track-aware contrastive learning; and prediction-level consistency stabilizes fusion through knowledge distillation. Extensive experiments demonstrate significant improvements under diverse audio, visual and joint corruptions, while maintaining competitive performance on clean data.

View free PDFSource page

Related papers

arxivcs.CVcs.GRcs.HCcs.SD2026-06-29

SICAGE: Speaker-Independent Culture-Aware Gesture Generation using TED4C-L Dataset

Ariel Gjaci, Antonio Sgorbissa, Vittorio Murino

Recent co-speech gesture generation methods often overlook cultural differences, limiting their effectiveness in human-agent interaction. Moreover, culture-conditioned models are rarely evaluated under speaker-disjoint splits, so apparent "cultural" behavior may be confounded wit…

View free PDFSource page
arxivcs.CVcs.AIcs.MMcs.SD2026-07-16

SceneBind: Binding What and Where Across Vision, Audio and Language

Mingfei Chen, Zijun Cui, Ruoke Zhang, Hyeonggon Ryu, Eli Shlizerman

We present SceneBind, an omni-modal representation of realistic scenes with joint semantic and 3D spatial understanding across vision, audio and language. Existing omni-modal encoders excel at instance-level semantics (i.e., what is present), but often lack explicit spatial struc…

View free PDFSource page
arxivcs.CVcs.SD2026-07-05

UniSkip-Mamba: A Frequency-Aware State Space Model for Audio-Visual Temporal Forgery Localization

Cangjin Qiu, Quan Zhang, Dan Jiang, Ke Zhang

With the proliferation of AI-generated content, sophisticated multimedia manipulation has raised critical concerns about malicious applications such as opinion manipulation and evidence fabrication, making Audio-Visual Temporal Forgery Localization (AV-TFL) an urgent research fro…

View free PDFSource page
arxiveess.AScs.CVcs.SD2026-06-28

VIB-AVSR: Variational Information Bottleneck for Noise-Robust LLM-Based Audio-Visual Speech Recognition

Piyush Arora, Navlika Singh, Umberto Cappellazzo, Stavros Petridis, Maja Pantic

Audio-Visual Speech Recognition takes two input modalities, acoustic and visual streams, where visual information from lip movements aids recognition when audio is noisy. Recently, LLM-based AVSR models have emerged as a promising paradigm by connecting pre-trained audio-visual e…

View free PDFSource page
arxivcs.CVcs.MMcs.SDeess.ASeess.IV2026-07-15

Bring Music The Horizon: Music-Driven 360$^\circ$ Video Generation

Kai Hsu Tsai, Yong Wei Fu, Hung I Yang, Yu-Chih Chen

Music visualization offers a powerful way to enhance listeners' understanding and experience of music by translating auditory signals into visual forms. However, most existing approaches either rely heavily on lyrics or generate flat, non-immersive videos similar to conventional…

View free PDFSource page
arxivcs.CVcs.SD2026-07-15

MultiRef-Compass: Towards Comprehensive Evaluation of Multi-Reference-to-Audio-Video Generation

Xiaohan Zhang, Yuqing Wen, Junlin Chen, Yuqi Tang, Yiting He, Lizhuo Shao, et al.

Multi-reference-to-audio-video (MR2AV) generation aims to generate coherent audio-video content conditioned on multiple references and textual instructions. Existing benchmarks mainly focus on text-driven generation, single-reference subject preservation, or isolated audio-video…

View free PDFSource page