DeepLearning ru:
Clockwork Convnets for Video Semantic Segmentation.

Adaptive video processing by incorporating data-driven clocks.

We define a novel family of "clockwork" convnets driven by fixed or adaptive clock signals that schedule the processing of different layers at different update rates according to their semantic stability. We design a pipeline schedule to reduce latency for real-time recognition and a fixed-rate schedule to reduce overall computation. Finally, we extend clockwork scheduling to adaptive video processing by incorporating data-driven clocks that can be tuned on unlabeled video.

https://arxiv.org/pdf/1608.03609v1.pdf
https://github.com/shelhamer/clockwork-fcn

http://www.gitxiv.com/posts/89zR7ATtd729JEJAg/clockwork-convnets-for-video-semantic-segmentation

#dl #CV #Caffe #video #Segmentation

There is a new $1MM competition on Kaggle to use ML / AI to diagnose lung cancer from CT scans.

Not only it is the great breakthrough for Kaggle (it is the first competition with this huge prize fund), it is also a breakthrough for science, since top world researchers and enginners will compete to basically crowdsource and ease the lung cancer diagnostics.

Competition is available at: https://www.kaggle.com/c/data-science-bowl-2017

#kaggle #segmentation #deeplearning #cv

Google purchased scene segmentation technology.

https://techcrunch.com/2017/08/16/google-acquires-aimatter-maker-of-the-fabby-computer-vision-app/

#dl #segmentation #cv #google

Semantic Segmentation Models for Autonomous Vehicles


https://www.kdnuggets.com/2018/03/semantic-segmentation-models-autonomous-vehicles.html

#deeplearning #segmentation

ModaNet: A Large-Scale Street Fashion Dataset with Polygon Annotations

Latest segmentation and detection approaches (DeepLabV3+, FasterRCNN) applied to street fashion images. Arxiv paper contains information about both: net and dataset.

Arxiv link: https://arxiv.org/abs/1807.01394
Paperdoll dataset: http://vision.is.tohoku.ac.jp/~kyamagu/research/paperdoll/

#segmentation #dataset #fashion #sv

Cancer metastasis detection with neural conditional random field (NCRF)

Github: https://github.com/baidu-research/NCRF?utm_source=telegram&utm_medium=opendatascience

#Baidu #Cancer #Segmentation #cv #DL

Paper «A Probabilistic U-Net for Segmentation of Ambiguous Images» from #NIPS2018 spotlight presentation.

Github: https://github.com/SimonKohl/probabilistic_unet
Github: Arxiv: https://arxiv.org/abs/1806.05034

#DeepMind #segmentation #cv

Faster R-CNN and Mask R-CNN in #PyTorch 1.0

Another release from #Facebook.

Mask R-CNN Benchmark: a fast and modular implementation for Faster R-CNN and Mask R-CNN written entirely in @PyTorch 1.0. It brings up to 30% speedup compared to mmdetection during training.

Webcam demo and ipynb file are available.

Github: https://github.com/facebookresearch/maskrcnn-benchmark

#CNN #CV #segmentation #detection

​​Papers from #DeepMind panel at #NIPS2018

Work on radiotherapy planning: https://arxiv.org/abs/1809.04430
Triaging eye diseases: https://www.nature.com/articles/s41591-018-0107-6
Probabilistic U-net: https://arxiv.org/abs/1806.05034

#segmentation #CV #Unet

​​Fast video object segmentation with Spatio-Temporal GANs

Spatio-Temporal GANs to the Video Object Segmentation task, allowing to run at 32 FPS without fine-tuning.

#FaSTGAN #GAN #Segmentation #videomining #CV #DL

​​Deep Learning Image Segmentation for Ecommerce Catalogue Visual Search

Microsoft’s article on image segmentation

Link: https://www.microsoft.com/developerblog/2018/04/18/deep-learning-image-segmentation-for-ecommerce-catalogue-visual-search/

#CV #DL #Segmentation #Microsoft

​​GSCNN: video segmetation architecture

Semantic segmentation GSCNN significantly outperforms DeepLabV3+ on Cityscapes benchmark.

Paper: https://arxiv.org/abs/1907.05740
Github (Project): https://github.com/nv-tlabs/GSCNN

#DL #CV #NVidiaAI #Nvidia #autonomous #selfdriving #car #RL #segmentation

​​New paper on training with pseudo-labels for semantic segmentation

Semi-Supervised Segmentation of Salt Bodies in Seismic Images:
SOTA (1st place) at TGS Salt Identification Challenge.

Github: https://github.com/ybabakhin/kaggle_salt_bes_phalanx
ArXiV: https://arxiv.org/abs/1904.04445

#GCPR2019 #Segmentation #CV

Video on how Facebook continues to develop its #Portal device

How #Facebook used Mask R-CNN, #PyTorch, and custom hardware integrations like foveated processing to improve Portal’s Smart Camera system.

Link: https://ai.facebook.com/blog/smart-camera-portal-advances/

#CV #DL #Segmentation

YOLACT: Real-time Instance Segmentation

Fully-convolutional model for real-time instance segmentation that achieves 29.8 mAP on MS COCO at 33.5 fps evaluated on a single Titan Xp, which is significantly faster than any previous competitive approach. They obtain this result after training on only one GPU.


video: https://www.youtube.com/watch?v=0pMfmo8qfpQ
paper: https://arxiv.org/abs/1904.02689
code: https://github.com/dbolya/yolact

#yolo #instance_segmentation #segmentation #real_time

BodyPix: Real-time Person Segmentation in the Browser with TensorFlow.js

Released BodyPix 2.0 with #multiperson support and improved accuracy (based on ResNet50), a new API, weight quantization, and support for different image sizes with TensorFlow.js

It estimates and renders person and body-part segmentation at 25 fps on a 2018 15-inch MacBook Pro, and 21 fps on an iPhone X.

code: https://github.com/tensorflow/tfjs-models/tree/master/body-pix
demo: https://storage.googleapis.com/tfjs-models/demos/body-pix/index.html
blog: https://blog.tensorflow.org/2019/11/updated-bodypix-2.html

#dl #segmentation

​​SAUNet: Shape Attentive U-Net for Interpretable Medical Image Segmentation

New approach for interpreting medical image segmentation models.

U-Net and other image segmentation models work quite well on medical data, but still aren't widely adopted. One of the reasons is the lack of reproducibility as well as robustness issues.
The key idea of the paper is using the additional stream in U-Net with shape features to increase robustness and use the output of this stream (attention map) that can be used or interpretability.

Modifications to the basic U-Net architecture:
- use dense blocks from DenseNet-121 as the encoder.
- use dual attention decoder block (with spatial and channel-wise attention paths)
- make the second stream using object shape (contour)
- dual-task loss function: cross-entropy + dice + edge loss (bce loss of the predicted shape boundaries)

Shape and spatial attention maps can be used for interpretation.

Paper: https://arxiv.org/abs/2001.07645
Code: https://github.com/sunjesse/shape-attentive-unet


#unet #imagesegmentation #interpretability #segmentation