SimMIM
Unsupervised-Classification
| SimMIM | Unsupervised-Classification | |
|---|---|---|
| 1 | 2 | |
| 1,030 | 1,456 | |
| 0.0% | 0.0% | |
| 0.0 | 1.4 | |
| over 3 years ago | almost 3 years ago | |
| Python | Python | |
| MIT License | GNU General Public License v3.0 or later |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
SimMIM
-
My DL model mis-segment wrong objects with similar colors
It seems the segmentation task is complex. One thing that I can suggest that definitely worked for me is Self-supervised learning prior to training. The task of the sel-supervision is very important. In your case I suggest Masked Image Modeling because it does not rely heavily on shapes and colors but can leverage them if they exist. Check out SimMIM by Microsoft. https://arxiv.org/abs/2111.09886 . https://github.com/microsoft/simmim
Unsupervised-Classification
-
Middle ground dataset between CIFAR and ImageNet [D]
The subsets we used are from here: https://github.com/wvangansbeke/Unsupervised-Classification/tree/master/data/imagenet_subsets
- Any reference or idea about how to train unsupervised CNN model ?
What are some alternatives?
SparK - [ICLR'23 Spotlightš„] The first successful BERT/MAE-style pretraining on any convolutional network; Pytorch impl. of "Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling"
s3prl - Self-Supervised Speech Pre-training and Representation Learning Toolkit
PaddleClas - A treasure chest for visual classification and recognition powered by PaddlePaddle
Unsupervised-Semantic-Segmentation - Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals. [ICCV 2021]
Swin-Transformer - This is an official implementation for "Swin Transformer: Hierarchical Vision Transformer using Shifted Windows".
contrastive-reconstruction - Tensorflow-keras implementation for Contrastive Reconstruction (ConRec) : a self-supervised learning algorithm that obtains image representations by jointly optimizing a contrastive and a self-reconstruction loss.