- VConv-DAE: Deep Volumetric Shape Learning Without Object Labels (ECCV-W’16)
- Deep Reflectance Maps (CVPR’16)
- DeLight-Net: Decomposing Reflectance Maps into Specular Materials and Natural Illumination (ArXiv’16)
- Novel Views of Objects from a Single Image (ArXiv’16)
- Visual Turing Challenge (NIPS’14, ICCV’15)
- Person Reidentification (ICCV’15)
- Sequential Bayesian Model Update under Structured Scene Prior for Semantic Road Scenes Labeling (ICCV’13)
- Learning Smooth Pooling Regions for Visual Recognition (BMVC’13)
- Multi-Class Video Co-Segmentation (CVPR’13)
- Rendered Materials (ECCV’12)