• Publications
  • Influence
Deep Learning Face Attributes in the Wild
TLDR
We propose a novel deep learning framework for attribute prediction in the wild, which outperforms the state-of-the-art with a large margin. Expand
  • 2,777
  • 745
  • PDF
Dynamic Graph CNN for Learning on Point Clouds
TLDR
We propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds, including classification and segmentation. Expand
  • 783
  • 226
  • PDF
DeepFashion: Powering Robust Clothes Recognition and Retrieval with Rich Annotations
TLDR
We introduce DeepFashion, a comprehensively annotated clothes dataset that contains massive attributes, clothing landmarks, as well as cross-pose/cross-domain correspondences of clothing pairs. Expand
  • 638
  • 183
  • PDF
Video Frame Synthesis Using Deep Voxel Flow
TLDR
We address the problem of synthesizing new video frames in an existing video, either in-between existing frames (interpolation), or subsequent to them (extrapolation). Expand
  • 312
  • 70
  • PDF
Hybrid Task Cascade for Instance Segmentation
TLDR
Cascade is a classic yet powerful architecture that has boosted performance on various tasks. Expand
  • 181
  • 36
  • PDF
Semantic Image Segmentation via Deep Parsing Network
TLDR
We solve MRF by proposing a Convolutional Neural Network (CNN), namely Deep Parsing Network (DPN), which enables deterministic end-to-end computation in a single forward pass. Expand
  • 472
  • 34
  • PDF
MMDetection: Open MMLab Detection Toolbox and Benchmark
TLDR
We present MMDetection, an object detection toolbox that contains a rich set of object detection and instance segmentation methods as well as related components and modules. Expand
  • 244
  • 26
  • PDF
Large-Scale Long-Tailed Recognition in an Open World
TLDR
We develop an integrated OLTR algorithm that maps an image to a feature space such that visual concepts can easily relate to each other based on a learned metric that respects closed-world classification while acknowledging the novelty of the open world. Expand
  • 86
  • 25
  • PDF
MaskGAN: Towards Diverse and Interactive Facial Image Manipulation
TLDR
We propose a novel framework termed MaskGAN, enabling diverse and interactive face manipulation with fidelity preservation. Expand
  • 80
  • 22
  • PDF
Fashion Landmark Detection in the Wild
TLDR
We present fashion landmark detection or fashion alignment, which is to predict the positions of functional key points defined on the fashion items, such as the corners of neckline, hemline, and cuff. Expand
  • 94
  • 20
  • PDF