Yutao Yang
DIRECT-ENTRY PH.D. STUDENT · ARTIFICIAL INTELLIGENCE
I am a direct-entry Ph.D. student in Electronic Information, jointly trained by Nankai University’s College of Excellent Engineers and Beijing Zhongguancun College. I received my B.Eng. in Software Engineering from Shanxi University in 2026.
My research aims to build intelligent systems that understand long-form visual dynamics, form predictive models of the world, and translate multimodal perception into structured actions. I am particularly interested in connecting video understanding with embodied agents and reliable, hierarchical decision-making.
News
Won First Prize in the Football Application Innovation Track at the First National Football AI Innovation Exchange Conference.
SeamSlicer was accepted by ACM Multimedia 2026 (CCF-A); I contributed as the second author.
Received the highest year-end performance tier during my Nankai University research internship.
Received direct-entry Ph.D. admission through graduate recommendation, ranked 4/801 (top 0.5%).
Research Interests
Video Understanding
Temporal modeling, multimodal event parsing, boundary detection, and long-form sports video analysis.
World Models
Learning predictive representations of environments for planning, reasoning, and embodied interaction.
Vision–Language–Action
Joint multimodal learning that grounds language and visual perception in executable actions.
Hierarchical Planning
Task decomposition and structured decision-making for reliable long-horizon intelligent agents.
Selected Publication
Speech-Aware Multimodal Video Slicing for Automated Video Trimming
Yandong Liu, Yutao Yang, Shengjiao Dong, Wenqiang Zhang, Xiang Li, Mengli Yu, Zhao-Min Chen, Lingfeng Yang
Proceedings of the 34th ACM International Conference on Multimedia (ACM MM 2026) · Accepted · CCF-A
Introduces SeamSlicer, a multimodal framework for atomic-level video boundary detection and semantically complete automated trimming, together with the AtomicBench benchmark.
Research & Projects
LumenSport
An automated sports-video analysis and editing system for football and basketball. I designed the end-to-end pipeline for event understanding, highlight localization, temporal segmentation, and automatic clip generation, integrating visual, audio, and match-context cues. The project supported our First Prize result in the national Football AI innovation competition.
SeamSlicer & AtomicBench
Studied atomic-level automated trimming by jointly modeling visual, speech, text, music, sound effects, and ambient audio. Contributed to the three-stage boundary-extraction, visual-refinement, and length-control framework and to AtomicBench construction and evaluation.
Siamese Network for Surveillance Anomaly Detection
Built a shared-weight Siamese visual-similarity baseline and a four-branch metric-learning strategy, combining efficient convolution, multi-scale feature fusion, skip connections, and learnable pair/triplet-loss weighting for illumination and occlusion robustness.
Fire & Smoke Detection
Built a YOLOv8-based fire-and-smoke detector with parallel dilated-convolution context enhancement, MCA attention, and a coordinate-attention-guided hierarchical feature-selection pyramid, achieving 91.8% mAP on the project dataset.
MakeupVision
Led a virtual-makeup vision pipeline integrating Light-FaceDetector, 106-point landmark estimation, and facial semantic segmentation; developed ROI localization, lip/eye masks, and cross-layer texture transfer for localized makeup synthesis.
Education
Nankai University · Beijing Zhongguancun College
Direct-entry Ph.D. in Electronic Information, jointly trained through the College of Excellent Engineers.
Shanxi University
B.Eng. in Software Engineering · Recommendation ranking 4/801 (top 0.5%) · Comprehensive ranking 7/801 · GPA 3.74/4.50.
Honors & Awards
National Competitions
First National Football AI Innovation Exchange Conference, Football Application Innovation Track
National 3D Digital Innovation Design Competition, 17th Anniversary Elite League
National College Business Elite Challenge, Brand Planning National Finals
16th Lanqiao Cup National Finals, Python Programming
2025 Group Programming Ladder Tournament National Finals
View provincial and university competitions
Provincial Competitions
National 3D Digital Innovation Design Competition, Shanxi Division
China International College Students’ Innovation Competition, Shanxi Division
16th Lanqiao Cup, Python Programming, Shanxi Division
Group Programming Ladder Tournament, Shanxi Division
Chinese Collegiate Computing Competition, Digital Media Game & Interaction
15th Lanqiao Cup, Python Programming, Shanxi Division
15th Chinese Mathematics Competition, Non-Mathematics Group A
16th Lanqiao Cup Visual Art Design, Static Poster
2024 National Undergraduate Mathematical Contest in Modeling, Shanxi
17th Chinese Collegiate Computing Competition, Software Applications and Development
14th Lanqiao Cup, C/C++ Programming, Shanxi Division
27th China Robotics and Artificial Intelligence Competition, Provincial Selection
National Undergraduate Mathematical Contest in Modeling, Shanxi
University Competitions
15th Market Survey and Analysis Competition, Shanxi University Selection
China International College Students’ Innovation Competition, Shanxi University
10th New-Form Software Competition, 3D Digital Art Teaching System
Scholarships & Distinctions
Outstanding Undergraduate Graduate, Shanxi University
National Encouragement Scholarship (three consecutive awards)
First-Class Academic Scholarship (three consecutive awards)
Three-Good Student, Shanxi University
Highest Year-End Performance Tier, Nankai University Research Internship
Skills & Credentials
Programming
Python · C/C++ · Java · MATLAB · SQL · JavaScript/TypeScript
AI & Vision
PyTorch · OpenCV · NumPy · Pandas · scikit-learn · multimodal learning · temporal modeling
Engineering
Linux · Git · React · Vite · LaTeX · Java Web · Unity · 3ds Max · SolidWorks · SPSS
Credentials
Huawei HCIA · CET-4 & CET-6 · Kaggle Stanford RNA 3D Folding: 331/1516
Leadership, Service & Practice
Student leadership. Organized calligraphy exhibitions, member activities, and publicity for the Shanxi University Calligraphy and Painting Association. Also served as class safety and mental-health committee member.
Community service. Completed 202 hours of in-person volunteering during undergraduate study.
Technical writing. Published practical AI and software-engineering articles on CSDN, with more than 420,000 total views and 13,000 followers.
Calligraphy. Practice Chinese brush calligraphy, especially Slender Gold script (瘦金体) and regular script (楷书).