Deeplab-resnet-101 in Pytorch with Jaccard loss

Related tags

Deep Learning jaccardSegment
Overview

Deeplab-resnet-101 Pytorch with Lovász hinge loss

Train deeplab-resnet-101 with binary Jaccard loss surrogate, the Lovász hinge, as described in http://arxiv.org/abs/1705.08790.

Parts of the code is adapted from tensorflow-deeplab-resnet (in particular the conversion from caffe to tensorflow with kaffe).

The code has not been tested for full training of Deeplab-Resnet yet. Refer to tensorflow-deeplab-resnet and possibly extract the weights after training with that framework.

Code status

The code is in early stage. Pull requests welcome.

Citation

Please cite

@ARTICLE{2017arXiv170508790B,
   author = {{Berman}, M. and {Blaschko}, M.~B.},
    title = "{Optimization of the Jaccard index for image segmentation with the Lov\'asz hinge}",
  journal = {ArXiv e-prints},
archivePrefix = "arXiv",
   eprint = {1705.08790},
 primaryClass = "cs.CV",
 keywords = {Computer Science - Computer Vision and Pattern Recognition},
     year = 2017,
    month = may,
   adsurl = {http://adsabs.harvard.edu/abs/2017arXiv170508790B},
}

if you use the code.

Dependencies and weights

Relies notably on Pytorch and the standalone tensorboard package

Using anaconda, install the full requirements using the provided conda environment file:

conda env create --f environemnt.yml
source activate jaccard-segment

Convert the Deeplab Caffe weights to tensorflow ckpt using caffe-tensorflow, then convert them to hdf5 using ckpt_to_dd.py and use our wrapper to load in Pytorch.

Important switches in the settings

By default, finetunes with cross-entropy loss. Use --binary class switch for selecting a particular class in the binary case, --jaccard for training with the Jaccard hinge loss described in the arxiv paper, --hinge to use the Hinge loss, and --proximal to use the prox. operator optimization variant for the Jaccard loss as described in the arxiv paper.

For the prox. operator, use a learning rate of 1. and set an equivalent regularization of 1/lr instead.

You might also like...
Pretrained models for Jax/Flax: StyleGAN2, GPT2, VGG, ResNet.
Pretrained models for Jax/Flax: StyleGAN2, GPT2, VGG, ResNet.

Pretrained models for Jax/Flax: StyleGAN2, GPT2, VGG, ResNet.

Matthias Wright 169 Dec 26, 2022
improvement of CLIP features over the traditional resnet features on the visual question answering, image captioning, navigation and visual entailment tasks.

CLIP-ViL In our paper "How Much Can CLIP Benefit Vision-and-Language Tasks?", we show the improvement of CLIP features over the traditional resnet fea

null 310 Dec 28, 2022
Reproduce ResNet-v2(Identity Mappings in Deep Residual Networks) with MXNet
Reproduce ResNet-v2(Identity Mappings in Deep Residual Networks) with MXNet

Reproduce ResNet-v2 using MXNet Requirements Install MXNet on a machine with CUDA GPU, and it's better also installed with cuDNN v5 Please fix the ran

Wei Wu 531 Dec 4, 2022
NFT-Price-Prediction-CNN - Using visual feature extraction, prices of NFTs are predicted via CNN (Alexnet and Resnet) architectures.

NFT-Price-Prediction-CNN - Using visual feature extraction, prices of NFTs are predicted via CNN (Alexnet and Resnet) architectures.

null 5 Nov 3, 2022
In this project we use both Resnet and Self-attention layer for cat, dog and flower classification.
In this project we use both Resnet and Self-attention layer for cat, dog and flower classification.

cdf_att_classification classes = {0: 'cat', 1: 'dog', 2: 'flower'} In this project we use both Resnet and Self-attention layer for cdf-Classification.

null 3 Nov 23, 2022
Pretrained models for Jax/Haiku; MobileNet, ResNet, VGG, Xception.

Pre-trained image classification models for Jax/Haiku Jax/Haiku Applications are deep learning models that are made available alongside pre-trained we

Alper Baris CELIK 14 Dec 20, 2022
Official implementation of our paper
Official implementation of our paper "LLA: Loss-aware Label Assignment for Dense Pedestrian Detection" in Pytorch.

LLA: Loss-aware Label Assignment for Dense Pedestrian Detection This project provides an implementation for "LLA: Loss-aware Label Assignment for Dens

null 35 Dec 6, 2022
PyTorch implementation of Soft-DTW: a Differentiable Loss Function for Time-Series in CUDA
PyTorch implementation of Soft-DTW: a Differentiable Loss Function for Time-Series in CUDA

Soft DTW Loss Function for PyTorch in CUDA This is a Pytorch Implementation of Soft-DTW: a Differentiable Loss Function for Time-Series which is batch

Keon Lee 76 Dec 20, 2022
PyTorch implementation for Partially View-aligned Representation Learning with Noise-robust Contrastive Loss (CVPR 2021)
PyTorch implementation for Partially View-aligned Representation Learning with Noise-robust Contrastive Loss (CVPR 2021)

2021-CVPR-MvCLN This repo contains the code and data of the following paper accepted by CVPR 2021 Partially View-aligned Representation Learning with

XLearning Group 33 Nov 1, 2022
Comments
  • Clarification for the greedy algorithm validity

    Clarification for the greedy algorithm validity

    Hello! Thank you for these fascinating research results. I guess this is a silly question, but still: Could you please provide an argument for the greediness in the algorithm used in the proximal operator computation? Precisely, what I am interested in is why, once reaching an edge of the polyhedron, we should move along that edge (not considering the other face) and why the edge direction is given by averaging the "clashed" components of the previous direction? The rest of the algorithm is well-grounded and clear to me. Thanks in advance!

    opened by kartynnik 0
Owner
Maxim Berman
Maxim Berman
GitHub
Models Supported: AlbUNet [18, 34, 50, 101, 152] (1D and 2D versions for Single and Multiclass Segmentation, Feature Extraction with supports for Deep Supervision and Guided Attention)

AlbUNet-1D-2D-Tensorflow-Keras This repository contains 1D and 2D Signal Segmentation Model Builder for AlbUNet and several of its variants developed

Sakib Mahmud 1 Nov 15, 2021
Implement of "Training deep neural networks via direct loss minimization" in PyTorch for 0-1 loss

This is the implementation of "Training deep neural networks via direct loss minimization" published at ICML 2016 in PyTorch. The implementation targe

Cuong Nguyen 1 Jan 18, 2022
Recall Loss for Semantic Segmentation (This repo implements the paper: Recall Loss for Semantic Segmentation)

Recall Loss for Semantic Segmentation (This repo implements the paper: Recall Loss for Semantic Segmentation) Download Synthia dataset The model uses

null 32 Sep 21, 2022
An implementation for the loss function proposed in Decoupled Contrastive Loss paper.

Decoupled-Contrastive-Learning This repository is an implementation for the loss function proposed in Decoupled Contrastive Loss paper. Requirements P

Ramin Nakhli 71 Dec 4, 2022
PyTorch implementation of the R2Plus1D convolution based ResNet architecture described in the paper "A Closer Look at Spatiotemporal Convolutions for Action Recognition"

R2Plus1D-PyTorch PyTorch implementation of the R2Plus1D convolution based ResNet architecture described in the paper "A Closer Look at Spatiotemporal

Irhum Shafkat 342 Dec 16, 2022
PyTorch implementation of MoCo v3 for self-supervised ResNet and ViT.

MoCo v3 for Self-supervised ResNet and ViT Introduction This is a PyTorch implementation of MoCo v3 for self-supervised ResNet and ViT. The original M

Facebook Research 887 Jan 8, 2023
Base pretrained models and datasets in pytorch (MNIST, SVHN, CIFAR10, CIFAR100, STL10, AlexNet, VGG16, VGG19, ResNet, Inception, SqueezeNet)

This is a playground for pytorch beginners, which contains predefined models on popular dataset. Currently we support mnist, svhn cifar10, cifar100 st

Aaron Chen 2.4k Dec 28, 2022
Reproduces ResNet-V3 with pytorch

ResNeXt.pytorch Reproduces ResNet-V3 (Aggregated Residual Transformations for Deep Neural Networks) with pytorch. Tried on pytorch 1.6 Trains on Cifar

Pau Rodriguez 481 Dec 23, 2022
3D ResNet Video Classification accelerated by TensorRT

Activity Recognition TensorRT Perform video classification using 3D ResNets trained on Kinetics-400 dataset and accelerated with TensorRT P.S Click on

Akash James 39 Nov 21, 2022
Quickly comparing your image classification models with the state-of-the-art models (such as DenseNet, ResNet, ...)

Image Classification Project Killer in PyTorch This repo is designed for those who want to start their experiments two days before the deadline and ki

null 349 Dec 8, 2022