A system for quickly generating training data with weak supervision

Pretrained DeepLabv3 and DeepLabv3+ for Pascal VOC & Cityscapes

Implementation of Attention-based Deep Multiple Instance Learning in PyTorch

[ICLR 2021, Spotlight] Large Scale Image Completion via Co-Modulated Generative Adversarial Networks

PyTorch implementation of Trust Region Policy Optimization

Code for our ACMMM2020 paper "Context-aware Feature Generation for Zero-shot Semantic Segmentation".

[AAAI 2022] The first dataset on foreground object shadow generation for image composition in real-world scenes. The code used in our paper "Shadow Generation for Composite Image in Real-world Scen…

[CVPR 2022] We unify pixel-to-pixel transformation and color-to-color transformation in a coherent framework for high-resolution image harmonization. We also release 100 high-resolution real compos…

[ECCV 2022] Official code for "Learning Object Placement via Dual-path Graph Completion"

The first dataset of composite images with rationality score indicating whether the object placement in a composite image is reasonable.

[IJCAI 2022] The first public benchmark dataset for video harmonization. The code used in our paper "Deep Video Harmonization with Color Mapping Consistency", IJCAI 2022.

[ECCV 2022] Generate sample-specific intra-category deltas for few-shot image generation.

Start from Interpret Neural Networks by Identifying Critical Data Routing Paths