Guangyao Su
Research Subjects
Guangyao Su has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Tao Zhou
China Unicom Software Research Institute, Beijing, China
4 shared publications - Bin Wu
Beijing University of Posts and Telecommunications, Beijing, Beijing, CHINA
3 shared publications - Jianwei Fang
China Unicom Software Research Institute, Beijing, China
2 shared publications - Bo Wang
Beijing University of Posts and Telecommunications, Beijing, Beijing, CHINA
1 shared publication - Chunjia Wang
China Unicom, Beijing, China
1 shared publication - Chunping Zheng
Beijing University of Posts and Telecommunications
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 17899-17906
Spectral-Prior Asymmetric Fusion with NLL-Guided Ensemble Learning for EEG–fNIRS Imagined Handwriting Decoding
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 17899-17906
Open AccessSpectral-Prior Asymmetric Fusion with NLL-Guided Ensemble Learning for EEG–fNIRS Imagined Handwriting Decoding
- Open Access
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1317-1326
DenseSpeech: Dense Multi-Segment Temporal Grounding in Public Speaking Videos
Temporal video grounding aims to localize temporal segments in untrimmed videos based on natural language queries. While dense grounding methods accept paragraph queries, they retain a restrictive one-to-one mapping between sentences and segments. This assumption fails in real-world behavioral analysis like public speaking, where a single query often corresponds to multiple, scattered, and overlapping segments due to event repetition, co-occurrence, and cross-modal complexity. To address this gap, we extend dense video grounding to a multi-segment setting and present DenseSpeech, the first densely annotated video grounding dataset in the speech domain. DenseSpeech comprises 1,800 authentic classroom recordings with 13,880 annotated segments spanning over 52 hours, featuring high annotation density, frequent boundary overlap, and pervasive event repetition with form variability. We further propose DenseMSG, an end-to-end dense multi-scale grounding model. Its Cross-modal Consistent Interaction module employs text-guided cross-attention to highlight query-relevant temporal regions and performs hierarchical bidirectional audiovisual fusion with adaptive gating to build a multi-scale feature pyramid. A Multi-Scale Integration module enables bidirectional feature propagation across scales for complementary global-local modeling, while similarity-based classification and query-aware regression heads jointly predict boundaries at all scales, naturally supporting one-to-many retrieval. Experiments show DenseMSG achieves state-of-the-art performance on DenseSpeech and public benchmarks (TACoS, ActivityNet Captions), validating the dataset and model.
- 15 June 2026
- Published by Association for Computing Machinery (ACM)
- p. 1317-1326
Open AccessDenseSpeech: Dense Multi-Segment Temporal Grounding in Public Speaking Videos
Temporal video grounding aims to localize temporal segments in untrimmed videos based on natural language queries. While dense grounding methods accept paragraph queries, they retain a restrictive one-to-one mapping between sentences and segments. This assumption fails in real-world behavioral analysis like public speaking, where a single query often corresponds to multiple, scattered, and overlapping segments due to event repetition, co-occurrence, and cross-modal complexity. To address this gap, we extend dense video grounding to a multi-segment setting and present DenseSpeech, the first densely annotated video grounding dataset in the speech domain. DenseSpeech comprises 1,800 authentic classroom recordings with 13,880 annotated segments spanning over 52 hours, featuring high annotation density, frequent boundary overlap, and pervasive event repetition with form variability. We further propose DenseMSG, an end-to-end dense multi-scale grounding model. Its Cross-modal Consistent Interaction module employs text-guided cross-attention to highlight query-relevant temporal regions and performs hierarchical bidirectional audiovisual fusion with adaptive gating to build a multi-scale feature pyramid. A Multi-Scale Integration module enables bidirectional feature propagation across scales for complementary global-local modeling, while similarity-based classification and query-aware regression heads jointly predict boundaries at all scales, naturally supporting one-to-many retrieval. Experiments show DenseMSG achieves state-of-the-art performance on DenseSpeech and public benchmarks (TACoS, ActivityNet Captions), validating the dataset and model.
- Open Access
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 739-747
Cause and Effect: Video Social Relationship Recognition from Causal Perspective
- 27 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 739-747
Open AccessCause and Effect: Video Social Relationship Recognition from Causal Perspective
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1461-1469
Demonstration Meets Typed Events: Type Specific Video Semantic Role Labeling via Multimodal Prompting and Retrieval
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1461-1469
Open AccessDemonstration Meets Typed Events: Type Specific Video Semantic Role Labeling via Multimodal Prompting and Retrieval