Target-Aware Spatio-Temporal Reasoning via Answering Questions in Dynamics Audio-Visual Scenarios

1 code implementation • 21 May 2023 • Yuanyuan Jiang, Jianqin Yin

Recent works rely on elaborate target-agnostic parsing of audio-visual scenes for spatial grounding while mistreating audio and video as separate entities for temporal grounding.

Audio-visual Question Answering Audio-Visual Question Answering (AVQA) +3

Leveraging the Video-level Semantic Consistency of Event for Audio-visual Event Localization

1 code implementation • 11 Oct 2022 • Yuanyuan Jiang, Jianqin Yin, Yonghao Dang

In contrast to existing methods, we propose a novel video-level semantic consistency guidance network for the AVE localization task.

audio-visual event localization

A Unified and Fast Interpretable Model for Predictive Analytics

no code implementations • 16 Nov 2021 • Yuanyuan Jiang, Rui Ding, Tianchi Qiao, Yunan Zhu, Shi Han, Dongmei Zhang

Predictive analytics is human involved, thus the machine learning model is preferred to be interpretable.

Decision Making

ML4C: Seeing Causality Through Latent Vicinity

1 code implementation • NeurIPS 2021 • Haoyue Dai, Rui Ding, Yuanyuan Jiang, Shi Han, Dongmei Zhang

Starting from seeing that SCL is not better than random guessing if the learning target is non-identifiable a priori, we propose a two-phase paradigm for SCL by explicitly considering structure identifiability.