DeepLab is a state-of-art deep learning system for semantic image segmentation built on top of Caffe.

Ali

Last update: Nov 14, 2022

Related tags

Deep Learning DeepLab-Context

Overview

DeepLab

Introduction

DeepLab is a state-of-art deep learning system for semantic image segmentation built on top of Caffe.

It combines densely-computed deep convolutional neural network (CNN) responses with densely connected conditional random fields (CRF).

This distribution provides a publicly available implementation for the key model ingredients first reported in an arXiv paper, accepted in revised form as conference publication to the ICLR-2015 conference. It also contains implementations for methods supporting model learning using only weakly labeled examples, described in a second follow-up arXiv paper. Please consult and consider citing the following papers:

@inproceedings{chen14semantic,
  title={Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs},
  author={Liang-Chieh Chen and George Papandreou and Iasonas Kokkinos and Kevin Murphy and Alan L Yuille},
  booktitle={ICLR},
  url={http://arxiv.org/abs/1412.7062},
  year={2015}
}

@article{papandreou15weak,
  title={Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation},
  author={George Papandreou and Liang-Chieh Chen and Kevin Murphy and Alan L Yuille},
  journal={arxiv:1502.02734},
  year={2015}
}

Note that if you use the densecrf implementation, please consult and cite the following paper:

@inproceedings{KrahenbuhlK11,
  title={Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials},
  author={Philipp Kr{\"{a}}henb{\"{u}}hl and Vladlen Koltun},
  booktitle={NIPS},      
  year={2011}
}

Performance

DeepLab currently achieves 73.9% on the challenging PASCAL VOC 2012 image segmentation task -- see the leaderboard.

Pre-trained models

We have released several trained models and corresponding prototxt files at here. Please check it for more model details.

The best model among the released ones yields 73.6% on PASCAL VOC 2012 test set.

Python wrapper requirements

Install wget library for python

sudo pip install wget

Change DATA_ROOT to point to the PASCAL images
To use the mat_read_layer and mat_write_layer, please download and install matio.

Running the code

python run.py

FAQ

Check FAQ if you have some problems while using the code.

Comments

DeepLabV2-ResNet101 - loss going up during training

I set up the DeepLabV2-ResNet101 model from the bitbucket code (so not using this python implementation) and during training I notice that the loss is going up. As I run this program for longer, the loss approaches 350. I was wondering if anyone would have any idea what could be the cause of this. Here is my run_pascal.sh script, I did not modify anything past the ## Training #1 (on train_aug) comment:

#/bin/sh
 
## MODIFY PATH for YOUR SETTING
ROOT_DIR=/deeplab
 
CAFFE_DIR=../code
CAFFE_BIN=${CAFFE_DIR}/.build_release/tools/caffe.bin
 
EXP=voc12
 
if [ "${EXP}" = "voc12" ]; then
    NUM_LABELS=21
    DATA_ROOT=${ROOT_DIR}/data/VOCdevkit/VOC2012
else
    NUM_LABELS=0
    echo "Wrong exp name"
fi
 
 
## Specify which model to train
########### voc12 ################
NET_ID=deeplabv2_resnet101
 
 
## Variables used for weakly or semi-supervisedly training
#TRAIN_SET_SUFFIX=
TRAIN_SET_SUFFIX=_aug
 
TRAIN_SET_STRONG=train
#TRAIN_SET_STRONG=train200
#TRAIN_SET_STRONG=train500
#TRAIN_SET_STRONG=train1000
#TRAIN_SET_STRONG=train750
 
TRAIN_SET_WEAK_LEN=0 #5000
 
DEV_ID=0
 
#####
 
## Create dirs
 
CONFIG_DIR=${EXP}/config/${NET_ID}
MODEL_DIR=${EXP}/model/${NET_ID}
mkdir -p ${MODEL_DIR}
LOG_DIR=${EXP}/log/${NET_ID}
mkdir -p ${LOG_DIR}
export GLOG_log_dir=${LOG_DIR}
 
## Run
 
RUN_TRAIN=1
RUN_TEST=0
RUN_TRAIN2=0
RUN_TEST2=0

Furthermore, I have the SegmentationClassAug ground truth images from your dropbox link (https://www.dropbox.com/s/oeu149j8qtbs1x0/SegmentationClassAug.zip?dl=0) in the data/VOCdevkit/VOC2012/ folder.

Here is what the training looks like:

I1127 01:09:18.232293 31928 net.cpp:270] This network produces output accuracy
I1127 01:09:18.232307 31928 net.cpp:270] This network produces output accuracy_res05
I1127 01:09:18.232322 31928 net.cpp:270] This network produces output accuracy_res075
I1127 01:09:18.232336 31928 net.cpp:270] This network produces output accuracy_res1
I1127 01:09:18.348408 31928 net.cpp:283] Network initialization done.
I1127 01:09:18.355520 31928 solver.cpp:60] Solver scaffolding done.
I1127 01:09:18.399452 31928 caffe.cpp:129] Finetuning from voc12/model/deeplabv2_resnet101/init.caffemodel
I1127 01:09:19.118443 31928 net.cpp:816] Ignoring source layer fc1_coco
I1127 01:09:19.118538 31928 net.cpp:816] Ignoring source layer fc1_coco_fc1_coco_0_split
I1127 01:09:19.132285 31928 caffe.cpp:219] Starting Optimization
I1127 01:09:19.132372 31928 solver.cpp:280] Solving deeplabv2_resnet101
I1127 01:09:19.132383 31928 solver.cpp:281] Learning Rate Policy: poly
I1127 01:09:25.117411 31928 solver.cpp:229] Iteration 0, loss = 261.803
I1127 01:09:25.117547 31928 solver.cpp:245]     Train net output #0: accuracy = 0.0291829
I1127 01:09:25.117589 31928 solver.cpp:245]     Train net output #1: accuracy = 0.0872093
I1127 01:09:25.117614 31928 solver.cpp:245]     Train net output #2: accuracy = 0.66929
I1127 01:09:25.117645 31928 solver.cpp:245]     Train net output #3: accuracy_res05 = 0.046683
I1127 01:09:25.117672 31928 solver.cpp:245]     Train net output #4: accuracy_res05 = 0.140741
I1127 01:09:25.117697 31928 solver.cpp:245]     Train net output #5: accuracy_res05 = 0.669494
I1127 01:09:25.117720 31928 solver.cpp:245]     Train net output #6: accuracy_res075 = 0.035668
I1127 01:09:25.117766 31928 solver.cpp:245]     Train net output #7: accuracy_res075 = 0.106589
I1127 01:09:25.117786 31928 solver.cpp:245]     Train net output #8: accuracy_res075 = 0.669965
I1127 01:09:25.117815 31928 solver.cpp:245]     Train net output #9: accuracy_res1 = 0.0486381
I1127 01:09:25.117841 31928 solver.cpp:245]     Train net output #10: accuracy_res1 = 0.145349
I1127 01:09:25.117866 31928 solver.cpp:245]     Train net output #11: accuracy_res1 = 0.627653
I1127 01:09:25.117985 31928 sgd_solver.cpp:106] Iteration 0, lr = 0.00025
I1127 01:10:54.876021 31928 solver.cpp:229] Iteration 20, loss = 336.749
I1127 01:10:54.877192 31928 solver.cpp:245]     Train net output #0: accuracy = 0
I1127 01:10:54.877221 31928 solver.cpp:245]     Train net output #1: accuracy = 0
I1127 01:10:54.877240 31928 solver.cpp:245]     Train net output #2: accuracy = 0.857143
I1127 01:10:54.877261 31928 solver.cpp:245]     Train net output #3: accuracy_res05 = 0
I1127 01:10:54.877290 31928 solver.cpp:245]     Train net output #4: accuracy_res05 = 0
I1127 01:10:54.877307 31928 solver.cpp:245]     Train net output #5: accuracy_res05 = 0.857143
I1127 01:10:54.877341 31928 solver.cpp:245]     Train net output #6: accuracy_res075 = 0
I1127 01:10:54.877368 31928 solver.cpp:245]     Train net output #7: accuracy_res075 = 0
I1127 01:10:54.877393 31928 solver.cpp:245]     Train net output #8: accuracy_res075 = 0.857143
I1127 01:10:54.877419 31928 solver.cpp:245]     Train net output #9: accuracy_res1 = 0
I1127 01:10:54.877435 31928 solver.cpp:245]     Train net output #10: accuracy_res1 = 0
I1127 01:10:54.877454 31928 solver.cpp:245]     Train net output #11: accuracy_res1 = 0.857143
I1127 01:10:54.877490 31928 sgd_solver.cpp:106] Iteration 20, lr = 0.000249775

opened by JustinLiang 11

Check failed: error == cudaSuccess (8 vs. 0) invalid device function, out of memory

I run the codes on Ubuntu 16.04, and caffe has GPU support (Quadro M4000, 8GB ), but one error occurs, anyone has ideas what's going on? how can I solve this problem? Many thanks for your help.

I0409 20:53:47.976828 21105 net.cpp:170] pool2 needs backward computation. I0409 20:53:47.976835 21105 net.cpp:170] relu2_2 needs backward computation. I0409 20:53:47.976843 21105 net.cpp:170] conv2_2 needs backward computation. I0409 20:53:47.976853 21105 net.cpp:170] relu2_1 needs backward computation. I0409 20:53:47.976861 21105 net.cpp:170] conv2_1 needs backward computation. I0409 20:53:47.976869 21105 net.cpp:170] pool1 needs backward computation. I0409 20:53:47.976878 21105 net.cpp:170] relu1_2 needs backward computation. I0409 20:53:47.976887 21105 net.cpp:170] conv1_2 needs backward computation. I0409 20:53:47.976897 21105 net.cpp:170] relu1_1 needs backward computation. I0409 20:53:47.976907 21105 net.cpp:170] conv1_1 needs backward computation. I0409 20:53:47.976915 21105 net.cpp:172] data does not need backward computation. I0409 20:53:47.976925 21105 net.cpp:208] This network produces output accuracy I0409 20:53:47.976969 21105 net.cpp:467] Collecting Learning Rate and Weight Decay. I0409 20:53:47.976986 21105 net.cpp:219] Network initialization done. I0409 20:53:47.976995 21105 net.cpp:220] Memory required for data: 9170170576 I0409 20:53:47.977139 21105 solver.cpp:41] Solver scaffolding done. I0409 20:53:47.977152 21105 caffe.cpp:118] Finetuning from voc12/model/vgg128_noup/init.caffemodel I0409 20:53:48.125612 21105 net.cpp:740] Target layer fc6 not initialized. I0409 20:53:48.125658 21105 net.cpp:740] Target layer fc7 not initialized. I0409 20:53:48.125666 21105 net.cpp:740] Target layer fc8_voc12 not initialized. I0409 20:53:48.127007 21105 solver.cpp:160] Solving vgg128_noup I0409 20:53:48.127018 21105 solver.cpp:161] Learning Rate Policy: step F0409 20:53:48.199013 21105 im2col.cu:68] Check failed: error == cudaSuccess (8 vs. 0) invalid device function *** Check failure stack trace: *** @ 0x7fd71f3715cd google::LogMessage::Fail() @ 0x7fd71f373433 google::LogMessage::SendToLog() @ 0x7fd71f37115b google::LogMessage::Flush() @ 0x7fd71f373e1e google::LogMessageFatal::~LogMessageFatal() @ 0x5c878a caffe::im2col_gpu<>() @ 0x5bd086 caffe::ConvolutionLayer<>::Forward_gpu() @ 0x55aaba caffe::Net<>::ForwardFromTo() @ 0x55aca7 caffe::Net<>::ForwardPrefilled() @ 0x5aef94 caffe::Solver<>::Solve() @ 0x41be0a train() @ 0x415708 main @ 0x7fd71c112830 __libc_start_main @ 0x41a769 _start @ (nil) (unknown) Aborted (core dumped)

opened by yghlc 3
Questions on softmax layer

Hi!

I am reproducing the results of DeepLab, and I met a problem.

I used the method mentioned here, and when I run run.py, it told me that there are unexpected labels in softmax layer (such as 55, 150 and so on). I have to force NUM_LABELS=255 to make it run.

How could I fix this problem? Thanks very much!

Shangxuan

opened by ShangxuanWu 3
Can not load the pre-trained models.

Hi Ali. Thanks for this program. I have one issue - can not load model: Downloading init caffemodel from http://ccvl.stat.ucla.edu/ccvl/init_models/vgg16_20M.caffemodel Traceback (most recent call last): File "run.py", line 67, in run(RUN_TRAIN, RUN_TEST, RUN_TRAIN2, RUN_TEST2, RUN_SAVE) File "run.py", line 55, in run if RUN_TRAIN : trainer() File "/home/lumius/Dropbox/DeepLab-Context/python/my_script/trainer.py", line 52, in trainer model = train_prototxt_maker(train, init, solver, train_set) File "/home/lumius/Dropbox/DeepLab-Context/python/my_script/trainer.py", line 32, in train_prototxt_maker if not os.path.isfile(model): model=model_finder(os.environ['EXP']+ '/model/' + os.environ['NET_ID']) File "/home/lumius/Dropbox/DeepLab-Context/python/my_script/tools.py", line 16, in model_finder filename = wget.download(url) File "/usr/local/lib/python2.7/dist-packages/wget.py", line 526, in download (tmpfile, headers) = ulib.urlretrieve(binurl, tmpfile, callback) File "/usr/lib/python2.7/urllib.py", line 98, in urlretrieve return opener.retrieve(url, filename, reporthook, data) File "/usr/lib/python2.7/urllib.py", line 245, in retrieve fp = self.open(url, data) File "/usr/lib/python2.7/urllib.py", line 213, in open return getattr(self, name)(url) File "/usr/lib/python2.7/urllib.py", line 364, in open_http return self.http_error(url, fp, errcode, errmsg, headers) File "/usr/lib/python2.7/urllib.py", line 377, in http_error result = method(url, fp, errcode, errmsg, headers) File "/usr/lib/python2.7/urllib.py", line 673, in http_error_301 return self.http_error_302(url, fp, errcode, errmsg, headers, data) File "/usr/lib/python2.7/urllib.py", line 642, in http_error_302 headers, data) File "/usr/lib/python2.7/urllib.py", line 669, in redirect_internal return self.open(newurl) File "/usr/lib/python2.7/urllib.py", line 213, in open return getattr(self, name)(url) File "/usr/lib/python2.7/urllib.py", line 443, in open_https h.endheaders(data) File "/usr/lib/python2.7/httplib.py", line 1053, in endheaders self._send_output(message_body) File "/usr/lib/python2.7/httplib.py", line 897, in _send_output self.send(msg) File "/usr/lib/python2.7/httplib.py", line 859, in send self.connect() File "/usr/lib/python2.7/httplib.py", line 1278, in connect server_hostname=server_hostname) File "/usr/lib/python2.7/ssl.py", line 353, in wrap_socket _context=self) File "/usr/lib/python2.7/ssl.py", line 601, in init self.do_handshake() File "/usr/lib/python2.7/ssl.py", line 830, in do_handshake self._sslobj.do_handshake() IOError: [Errno socket error] [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed (_ssl.c:590) then run run.py

opened by ghost 2
net->copytrainedlayer failed.

I have successfully compiled the code and trained a model using my own data.Then I wrote a c++ script to test a single image .But unfortunately, net->copytrainedlayer(model) seems not work ,weights of layers are still 0.Can you help me or just give me some clues ? ps, I highly doubt that if this function can not work well,the training process may be also incorrect.

I0413 11:50:28.412251 9144 net.cpp:170] relu1_2 needs backward computation. I0413 11:50:28.412251 9144 net.cpp:170] conv1_2 needs backward computation. I0413 11:50:28.413254 9144 net.cpp:170] relu1_1 needs backward computation. I0413 11:50:28.413254 9144 net.cpp:170] conv1_1 needs backward computation. I0413 11:50:28.414253 9144 net.cpp:172] data does not need backward computation. I0413 11:50:28.414253 9144 net.cpp:208] This network produces output accuracy I0413 11:50:28.414253 9144 net.cpp:467] Collecting Learning Rate and Weight Decay. I0413 11:50:28.416255 9144 net.cpp:219] Network initialization done. I0413 11:50:28.416255 9144 net.cpp:220] Memory required for data: 2445378832 I0413 11:50:28.417255 9144 solver.cpp:41] Solver scaffolding done. I0413 11:50:28.417255 9144 caffe.cpp:118] Finetuning from voc12/model/vgg128_noup/init.caffemodel I0413 11:50:28.417255 9144 net.cpp:740] Target layer conv1_1 not initialized. I0413 11:50:28.418267 9144 net.cpp:740] Target layer conv1_2 not initialized. I0413 11:50:28.418267 9144 net.cpp:740] Target layer conv2_1 not initialized. I0413 11:50:28.419257 9144 net.cpp:740] Target layer conv2_2 not initialized. I0413 11:50:28.419257 9144 net.cpp:740] Target layer conv3_1 not initialized. I0413 11:50:28.420267 9144 net.cpp:740] Target layer conv3_2 not initialized. I0413 11:50:28.420267 9144 net.cpp:740] Target layer conv3_3 not initialized. Could not create logging file: File exists COULD NOT CREATE A LOGGINGFILE 20170413-115028.4828!I0413 11:50:28.421259 9144 net.cpp:740] Target layer conv4_1 not initialized. I0413 11:50:28.422260 9144 net.cpp:740] Target layer conv4_2 not initialized. I0413 11:50:28.422260 9144 net.cpp:740] Target layer conv4_3 not initialized. I0413 11:50:28.422260 9144 net.cpp:740] Target layer conv5_1 not initialized. I0413 11:50:28.423260 9144 net.cpp:740] Target layer conv5_2 not initialized. I0413 11:50:28.423260 9144 net.cpp:740] Target layer conv5_3 not initialized. I0413 11:50:28.424262 9144 net.cpp:740] Target layer fc6 not initialized. I0413 11:50:28.424262 9144 net.cpp:740] Target layer fc7 not initialized. I0413 11:50:28.425261 9144 net.cpp:740] Target layer fc8_voc12 not initialized. I0413 11:50:28.426262 9144 solver.cpp:160] Solving vgg128_noup I0413 11:50:28.426262 9144 solver.cpp:161] Learning Rate Policy: step I0413 11:50:29.027125 9144 solver.cpp:209] Iteration 0, loss = 3.04452 I0413 11:50:35.966346 9144 solver.cpp:224] Train net output #0: accuracy = 0 I0413 11:50:35.967347 9144 solver.cpp:224] Train net output #1: accuracy = 0 I0413 11:50:35.967347 9144 solver.cpp:224] Train net output #2: accuracy = 0.571429

opened by yytzjgsu 2
How to pretrain model on MS-COCO?

I'm the new to semantic segmentation, the deeplab paper shows that pre-train models on coco dataset can much improve the results, but the coco dataset just give the instance level labels, how to transfer data to satisfy the category level requirement, or anyone can give me some examples on how to pre-train models on coco datasets? Thanks!

opened by yifan254 2
Reading .mat files

Hi,

I get .mat files as output but I don't have matlab. I heard we can use matio or scipy to read them but the shape of the array is like (579,579,21,1). I can't figure how to see them as images. Any help, pointers or suggestions are appreciated.

Thanks for sharing the wonderful project.

opened by codecolony 2
What do pointers to CAFFE_BIN and CAFFE_DIR mean?
In DeepLab-Context/python/my_script/tools.py lines 39-40

CAFFE_DIR='./' CAFFE_BIN='.build_release/tools/caffe.bin'

Do they just point to where caffe is installed in the system, or is it to create local directories considering it is a part of mkdir function?
opened by anuragranj 2
DeepLabV2-ResNet101 - Check failed: matfp Error creating MAT file voc12/features/deeplabv2_resnet101/test/fc1/2008_000006_blob_0.mat

I'm trying to run a test for the deeplabv2-resnet101 model but I ran into an error:

I commented out the accuracy layer (lines: 21825-21835) and uncommented the fc1_mat layer (lines: 21809-21823). Anyone know what the issue is?

opened by JustinLiang 1
shape mismatch at layer 'fc6'

Hi,

I'm getting the following error.

Cannot copy param 0 weights from layer 'fc6'; shape mismatch. Source param shape is 1024 512 3 3 (4718592); target param shape is 4096 512 4 4 (33554432). To learn this layer's parameters from scratch rather than copying from a saved net, rename the layer.

Any idea what is wrong here? Thanks in advance.

opened by codecolony 0
Code about Attention to Scale

Hi , I am interesting in the paper "Attention to Scale: Scale-aware Semantic Image Segmentation",you have given the address of the code ,but I can't find it . Would you please give me some help? Thanks a lot

opened by heng94 3
the version of caffe

hi, @TheLegendAli when I python run.py ,ERROR: [libprotobuf ERROR google/protobuf/text_format.cc:245] Error parsing text-format caffe.NetParameter: 24:3: Unknown enumeration value of "IMAGE_SEG_DATA" for field "type". It look like a version issue. my caffe version is 1.0.0, could you tell which version of yours. Or how do I get rid of this error? Help is greatly appreciated!

opened by cccusername 1
How to train on my data

HI @TheLegendAli

I'm try to train Deeplab v2 with my data. But i don't know how create data for training. In file train_aug.txt, only mapping between original image (JPEGImages/2007_000364.jpg) with contour image (/SegmentationClassAug/2007_000364.png) --> That seem all image in SegmentationClassAug folder is edge(contour). So how program can mapping color label for train?

Thanks

opened by ThienAnh 0

How to specify the validation data during training and output test accuracy

Hello, There is no test net accuracy in the log file even I specified the test data layer in the train.prototxt file:

layers {
  name: "data"
  type: IMAGE_SEG_DATA
  top: "data"
  top: "label"
  image_data_param {
    root_folder: ""
    source: "./list/test_aug.txt"
    label_type: PIXEL
    batch_size: 100
  }
  transform_param {
    mean_value: 75.209
    mean_value: 85.950
    mean_value: 95.685
    crop_size: 663
    mirror: false
  }
  include: { phase: TEST }
}

Only loss and train net accuracy in the log file, such as:

I0424 09:32:20.303715 32660 solver.cpp:209] Iteration 50, loss = 0.270298
I0424 09:32:20.303766 32660 solver.cpp:224]     Train net output #0: accuracy = 0.883158
I0424 09:32:20.303774 32660 solver.cpp:224]     Train net output #1: accuracy = 0.780627
I0424 09:32:20.303809 32660 solver.cpp:224]     Train net output #2: accuracy = 0.70205

How do I get the test accuracy during training?
What the meaning of Train net output #0, Train net output "#1", Train net output "#2"? "#0", "#1", "#2" means something?

Thanks for any suggestion!

opened by yghlc 1

Data augmentation has no effect?

when I add "scale_factors" in transform_param, I can not get better result,why?

transform_param { mean_value: 104.008 mean_value: 116.669 mean_value: 122.675 crop_size: 321 mirror: true scale_factors: 0.5 scale_factors: 0.75 scale_factors: 1 scale_factors: 1.25 scale_factors: 1.5 }

opened by NickMinYang 0

Owner

Ali

GitHub

AutoDeeplab / auto-deeplab / AutoML for semantic segmentation, implemented in Pytorch

AutoML for Image Semantic Segmentation Currently this repo contains the only working open-source implementation of Auto-Deeplab which, by the way out-

299 Dec 17, 2022

deep-table implements various state-of-the-art deep learning and self-supervised learning algorithms for tabular data using PyTorch.

63 Oct 17, 2022

State of the art Semantic Sentence Embeddings

Contrastive Tension State of the art Semantic Sentence Embeddings Published Paper · Huggingface Models · Report Bug Overview This is the official code

88 Dec 30, 2022

Caffe: a fast open framework for deep learning.

Caffe Caffe is a deep learning framework made with expression, speed, and modularity in mind. It is developed by Berkeley AI Research (BAIR)/The Berke

33k Dec 28, 2022

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.

MMdnn MMdnn is a comprehensive and cross-framework tool to convert, visualize and diagnose deep learning (DL) models. The "MM" stands for model manage

5.7k Jan 9, 2023

State of the Art Neural Networks for Deep Learning

pyradox This python library helps you with implementing various state of the art neural networks in a totally customizable fashion using Tensorflow 2

60 May 29, 2022

😇A pyTorch implementation of the DeepMoji model: state-of-the-art deep learning model for analyzing sentiment, emotion, sarcasm etc

------ Update September 2018 ------ It's been a year since TorchMoji and DeepMoji were released. We're trying to understand how it's being used such t

865 Dec 24, 2022

Deepparse is a state-of-the-art library for parsing multinational street addresses using deep learning

Here is deepparse. Deepparse is a state-of-the-art library for parsing multinational street addresses using deep learning. Use deepparse to Use the pr

192 Dec 20, 2022

Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation, CVPR 2018

Learning Pixel-level Semantic Affinity with Image-level Supervision This code is deprecated. Please see https://github.com/jiwoon-ahn/irn instead. Int

337 Dec 15, 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

32 Sep 21, 2022

Deep Text Search is an AI-powered multilingual text search and recommendation engine with state-of-the-art transformer-based multilingual text embedding (50+ languages).

Deep Text Search - AI Based Text Search & Recommendation System Deep Text Search is an AI-powered multilingual text search and recommendation engine w

19 Sep 29, 2022

DeepLab is a state-of-art deep learning system for semantic image segmentation built on top of Caffe.

Related tags

Overview

DeepLab

Introduction

Performance

Pre-trained models

Python wrapper requirements

Running the code

FAQ

Comments

Owner

Ali

AutoDeeplab / auto-deeplab / AutoML for semantic segmentation, implemented in Pytorch

deep-table implements various state-of-the-art deep learning and self-supervised learning algorithms for tabular data using PyTorch.

State of the art Semantic Sentence Embeddings

Caffe: a fast open framework for deep learning.

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.

State of the Art Neural Networks for Deep Learning

😇A pyTorch implementation of the DeepMoji model: state-of-the-art deep learning model for analyzing sentiment, emotion, sarcasm etc

Deepparse is a state-of-the-art library for parsing multinational street addresses using deep learning

Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation, CVPR 2018

Recall Loss for Semantic Segmentation (This repo implements the paper: Recall Loss for Semantic Segmentation)

Deep Text Search is an AI-powered multilingual text search and recommendation engine with state-of-the-art transformer-based multilingual text embedding (50+ languages).

DeepLab resnet v2 model in pytorch

Quickly comparing your image classification models with the state-of-the-art models (such as DenseNet, ResNet, ...)

TransFGU: A Top-down Approach to Fine-Grained Unsupervised Semantic Segmentation

Deep GPs built on top of TensorFlow/Keras and GPflow

QuickAI is a Python library that makes it extremely easy to experiment with state-of-the-art Machine Learning models.

Segmentation in Style: Unsupervised Semantic Image Segmentation with Stylegan and CLIP

hipCaffe: the HIP port of Caffe

a reimplementation of LiteFlowNet in PyTorch that matches the official Caffe version