-
GeoMVSNet: Learning Multi-View Stereo with Geometry Perception
CVPR 2023 · Paper · Supplementary · Demo · Video · Code -
Wan-Image: Pushing the Boundaries of Generative Visual Intelligence
arXiv, 2026 · Paper · Try Generation · API -
Points-to-3D: Bridging the Gap between Sparse Points and Shape-Controllable Text-to-3D Generation
ACM MM 2023 · Paper -
C³-GS: Learning Context-aware, Cross-dimension, Cross-scale Feature for Generalizable Gaussian Splatting
BMVC 2025 · Paper · Code -
FDC-NeRF: Learning Pose-free Neural Radiance Fields with Flow-Depth Consistency
ICASSP 2024 · Paper
-
📸 通义照相馆 | Tongyi Camera：人像写真图片生成 | Identity-Preserving AIGC
-
👗 通义虚拟试衣 | Tongyi Virtual Try-On：高保真虚拟试衣图片生成 | High-Fidelity Outfit VTON AIGC
-
🎥 万相视频生成 | Wan：领先的 AI 视频生成模型 | Leading AI Video Generation Model
-
👤 4D 写实数字人 | 4D Realistic Digital Human：4D 扫描写实数字人管线 | 4D scanning realistic digital human pipeline
-
🛕 MVS 三维重建 | MVS 3D Reconstruction：多视图立体三维重建 | Multi-view Stereo 3D reconstruction
-
🌙 二十二点零六 | Six Past Twenty-Two：微信小程序, “每个夜晚都会遇见” | A WeChat mini program about LOVE
-
🪕 遗音 | Lost Instrument：重新发现中国民族音乐 | Rediscover Chinese Folk Music
Email · Google Scholar · LinkedIn · Unsplash · 知乎