innerlee

Joined 6 years ago

Karma

8

Submissions

16

Member since

Aug 31

View on Hacker News

Recent submissions

Comment · 1 year ago

The singular defects (or high-norm tokens) [1] may be related to attention sinks. It is interesting that the direction of all high-norm tokens share the same direction. Maybe the theory behind is not very complex and the issue can be fixed cleverly during training.

[1] https://openreview.net/pdf?id=4yBnUokU2v

Comment · 5 years ago

For more text classification baselines (CRNN, NRTR, RubustScanner, SAR, SegOCR), checkout https://github.com/open-mmlab/mmocr They are reproducible, customizable.

Comment · 5 years ago

We also expored ways of presenting hundreds [1] of AI models in a better way. Recently we updated the model zoo so that it is arranged according to benchmark datasets [2] and papers [3]. Do you think it satisfies the needs?

[1] https://mmpose.readthedocs.io/en/latest/modelzoo.html [2] https://mmpose.readthedocs.io/en/latest/topics/body(2d,kpt,i... [3] https://mmpose.readthedocs.io/en/latest/papers/algorithms.ht...

Comment · 5 years ago

One of the maintainer here. We just now open-sourced a new ocr library. It contains 5 text detection algorithm, 4 text recognition algorithms, and 1 key information extraction algorithm (details see [1]). It is used internally for some time. We are exicted to join the OpenMMLab [2] ecosystem, and we hope you could try it, and let issues and prs coming!

[1] https://mmocr.readthedocs.io/en/latest/modelzoo.html [2] https://openmmlab.com/

OCR Baselines Based on PyTorch github.com

1 innerlee 5 years ago 1 comments

Comment · 5 years ago

Checkout the openmmlab project [1], where messy research codes in computer vision are rewritten an reorganized into a coherent whole. If there are more researchers join this, then not only research are fully reproducible, much more accessible to everyone, but also be compared fairly. (I'm the maintainer of mmpose [2], mmaction2 [3], and mmediting [4])

[1] https://github.com/open-mmlab [2] https://github.com/open-mmlab/mmpose [3] https://github.com/open-mmlab/mmaction2 [4] https://github.com/open-mmlab/mmediting

Comment · 5 years ago

Sure, checkpoints, configs and detailed training logs all are available at modelzoo https://mmaction2.readthedocs.io/en/latest/recognition_model...

The single RGB stream top1 goes up to 73.48% with resnet50, and up to 74.71% equipped with non-local. Both are much higher than the original paper with two-streams.

Comment · 5 years ago

shameless ad: try mmaction2, where every result is reproducible https://github.com/open-mmlab/mmaction2 . Modelzoo: https://mmaction2.readthedocs.io/en/latest/modelzoo.html

Tasks and Algorithms Supported by MMPose mmpose.readthedocs.io

1 innerlee 6 years ago 0 comments

MMAction2 v0.8.0 Released github.com

1 innerlee 6 years ago 0 comments

Comment · 6 years ago

Here is a demo video generated by mmpose: https://www.youtube.com/watch?v=pIJpQg8mXUU

MMPose: Pose Estimation in PyTorch github.com

2 innerlee 6 years ago 1 comments

Comment · 6 years ago

MMPose is an open-source toolbox for pose estimation based on PyTorch. It is a part of the OpenMMLab project. In V0.7.0, it adds support for the 3D human shape recovery task, adds support for WholeBody human pose estimation task, and the 2D hand keypoint estimation task.