Introduction
Y-tech Graphics AI Team at Kuaishou Technology was founded in 2019. Our team focuses on conducting cutting-edge research and development in the joint field of computer graphics, computer vision, and artificial intelligence, in order to empower hundreds of millions of Kuaishou users to create and consume content with innovative tools and technologies. Our team members are all experienced researchers located across different regions, including the headquarters in Beijing, the office in Hangzhou, and the Silicon Valley office in Palo Alto. We are working closely with other teams and have shipped several impactful features into the product line of Kuaishou.
Our current R&D interests include but are not limited to deep generative models, reinforcement learning, human digitization, photo-realistic rendering, appearance modeling, image & video editing/enhancement/stylization, as well as 3D reconstruction and scene understanding. Our team has published several papers in top-tier journals and conferences within a year, including ACM SIGGRAPH, IEEE TVCG, CVPR, ICCV, ECCV, ICML, and AAAI.
Selected Publications
- [ICCV 2021] Scene Synthesis via Uncertainty-Driven Attribute Synchronization
- [ICCV 2021] HPNet: Deep Primitive Segmentation Using Hybrid Representations
- [SIGGRAPH 2021] Volumetric Appearance Stylization with Stylizing Kernel Prediction Network
- [SIGGRAPH 2021] QuanTaichi: A Compiler for Quantized Simulations
- [CVPR 2021] Camera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D Registration
- [CVPR 2021] FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation
- [CVPR 2021] NBNet: Noise Basis Learning for Image Denoising with Subspace Projection
- [AAAI 2021] Arbitrary Video Style Transfer via Multi-Channel Correlation