Unverified50% confidenceBenchmarkExact time
The team constructed a benchmark dataset named BiomAF, containing paired expert-student videos, 3D skeletons, biomechanical attributes, and expert instructional annotations.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe research team designed a plug-and-play Biomechanical Data Parser (BDP) that converts raw video into structured biomechanical descriptors.75% similarUnverifiedHuman pose estimation technology can extract 3D skeletal motion sequences from 2D video, with mainstream solutions including OpenPose, MediaPipe, and MotionBERT.59% similarUnverifiedThe research team constructed a dataset called SCOPE-5K, containing approximately 5,000 video clips, covering both real and synthetic sources and involving multiple surgical scenarios such as gastrointestinal endoscopy and laparoscopy.59% similarUnverifiedTraining a CNN to recognize LEGO bricks requires massive amounts of labeled images, and similar industry projects typically use 3D modeling software to batch-render synthetic training sets of each brick type at various angles and lighting conditions.57% similarUnverified文生视频模型如Sora、Runway Gen-3、Kling本质上是在像素空间中进行概率性采样，基于训练数据统计规律生成画面57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983370API
curl https://kongchang.com/api/v1/knowledge/claims/983370MCP
get_claim(id=983370)