alibaba / Deep-Vision

Deep-Vision

This repository provides techniques developed by the Cognitive and Interactive Vision team from the City Brain Lab, Alibaba DAMO Academy.

The family of our works:

• Visual Analysis and Generation

  • Multiple Object Tracking with Correlation Learning (CVPR 2021)
  • Learning Position and Target Consistency for Memory-based Video Object Segmentation (CVPR2021)
  • RCL: Recurrent Continuous Localization for Temporal Action Detection (CVPR 2022)
  • Recurrent Dynamic Embedding for Video Object Segmentation (CVPR 2022)
  • Multi-View Consistent Generative Adversarial Networks for 3D-aware Image Synthesis (CVPR 2022)
  • Disentangled Representation Learning for Text-Video Retrieval (arXiv)
  • A Trend-Driven Fashion Design System for Rapid Response Marketing in E-commerce (AAAI 2022)
  • FastClip: An Efficient Video Understanding System with Heterogeneous Computing and Coarse-to-fine Processing (ACM Web 2022)
  • VideoDiscovery: An Automatic Short-Video Generation System for E-commerce Live-streaming (ACM-MM 2021)
  • Extremely Compact Non-local Representation Learning (KDD 2021)

• Self-supervised Learning

  • Self-supervised Video Representation Learning by Context and Motion Decoupling (CVPR 2021)
  • Train a One-Million-Way Instance Classifier for Unsupervised Visual Representation Learning (AAAI 2021)
  • Once and for All: Self-supervised Multi-modal Co-training on One-billion Videos at Alibaba (ACM-MM 2021)

• Efficient Deep Learning

  • Distribution Adaptive INT8 Quantization for Training CNNs (AAAI 2021)
  • Communication Efficient SGD via Gradient Sampling with Bayes Prior (CVPR 2021)
  • Accelerating Gossip SGD with Periodic Global Averaging (ICML 2021)
  • ANN Softmax: Acceleration of Extreme Classification Training (VLDB 2021)

Release

• Self-supervised Video Representation Learning by Context and Motion Decoupling
  [Project] [Paper] CVPR 2021
• Disentangled Representation Learning for Text-Video Retrieval
  [Project] [Paper]

License

This project is released under the Apache 2.0 license.

Copyright 2022 Alibaba Group Holding Ltd.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at the following link.

     http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.