Xiaohongshu Tech REDtech
Sep 21, 2023 · Artificial Intelligence
Xiaohongshu Team’s Four ICCV 2023 Papers on Open‑Vocabulary Video Instance Segmentation, One‑Shot 3D Avatar Learning, Test‑Time Personalized Human Pose Forecasting, and MPI‑Flow for Realistic Optical Flow
The Xiaohongshu technical team secured four ICCV 2023 papers—including an oral presentation—introducing an open‑vocabulary video instance segmentation benchmark and model, a one‑shot neural‑radiance‑field avatar method, a test‑time personalized 3D pose forecasting framework, and an MPI‑based realistic optical‑flow generation technique, all achieving state‑of‑the‑art performance.
3D avatarComputer VisionHuman Pose Forecasting
0 likes · 14 min read