EssentialMC2 is a complete system to solve video understanding tasks including MHRL(representation learning), MECR2(
relation reasoning) and MOSL3(openset life-long learning) powered by DAMO Academy
MinD(Machine IntelligenNce of Damo) Lab. This codebase provides a comprehensive solution for video classification,
temporal detection and noise learning.
Features
Simple and easy to use
High efficiency
Include SOTA papers presented by DAMO Academy
Include various pretrained models
Installation
Install by pip
Run pip install essmc2.
Install from source
Requirements
Python 3.6+
PytTorch 1.5+
Run python setup.py install. For each specific task, please refer to task specific README.
Model Zoo
Pretrained models can be found in the MODEL_ZOO.md.
SOTA Tasks
TAda! Temporally-Adaptive Convolutions for Video Understanding
[Project] [Paper] [Website] ICLR 2022
NGC: A Unified Framework for Learning with Open-World Noisy Data
[Project] [Paper] ICCV 2021
Self-supervised Motion Learning from Static Images
[Project] [Paper] CVPR 2021
A Stronger Baseline for Ego-Centric Action Detection
[Project] [Paper]
MIT License
Copyright (c) 2021 Alibaba
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
Acknowledgement
EssentialMC2 is an open source project that is contributed by researchers from DAMO Academy. We appreciate users who
give valuable feedbacks.
EssentialMC2
Introduction
EssentialMC2 is a complete system to solve video understanding tasks including MHRL(representation learning), MECR2( relation reasoning) and MOSL3(openset life-long learning) powered by DAMO Academy MinD(Machine IntelligenNce of Damo) Lab. This codebase provides a comprehensive solution for video classification, temporal detection and noise learning.
Features
Installation
Install by pip
Run
pip install essmc2.Install from source
Requirements
Run
python setup.py install. For each specific task, please refer to task specific README.Model Zoo
Pretrained models can be found in the MODEL_ZOO.md.
SOTA Tasks
[Project] [Paper] [Website] ICLR 2022
[Project] [Paper] ICCV 2021
[Project] [Paper] CVPR 2021
[Project] [Paper]
First-place submission to EPIC-KITCHENS-100 Action Detection Challenge
[Project] [Paper]
Second-place submission to EPIC-KITCHENS-100 Action Recognition challenge
License
EssentialMC2 is released under MIT license.
Acknowledgement
EssentialMC2 is an open source project that is contributed by researchers from DAMO Academy. We appreciate users who give valuable feedbacks.