Mask R-CNN
ICCV17 | 16 | Mask R-CNN
Kaiming He (Facebook AI Research), Georgia Gkioxari (Facebook), Piotr Dollar (Facebook AI Research), Ross Girshick (Facebook)
We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance. The method, called Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN, running at 5 fps. Moreover, Mask R-CNN is easy to generalize to other tasks, e.g., allowing us to estimate human poses in the same framework. We show top results in all three tracks of the COCO suite of challenges, including instance segmentation, bounding-box object detection, and person keypoint detection. Without tricks, Mask R-CNN outperforms all existing, single-model entries on every task, including the COCO 2016 challenge winners. We hope our simple and effective approach will serve as a solid baseline and help ease future research in instance-level recognition. Code will be made available.
Видео Mask R-CNN канала ComputerVisionFoundation Videos
Kaiming He (Facebook AI Research), Georgia Gkioxari (Facebook), Piotr Dollar (Facebook AI Research), Ross Girshick (Facebook)
We present a conceptually simple, flexible, and general framework for object instance segmentation. Our approach efficiently detects objects in an image while simultaneously generating a high-quality segmentation mask for each instance. The method, called Mask R-CNN, extends Faster R-CNN by adding a branch for predicting an object mask in parallel with the existing branch for bounding box recognition. Mask R-CNN is simple to train and adds only a small overhead to Faster R-CNN, running at 5 fps. Moreover, Mask R-CNN is easy to generalize to other tasks, e.g., allowing us to estimate human poses in the same framework. We show top results in all three tracks of the COCO suite of challenges, including instance segmentation, bounding-box object detection, and person keypoint detection. Without tricks, Mask R-CNN outperforms all existing, single-model entries on every task, including the COCO 2016 challenge winners. We hope our simple and effective approach will serve as a solid baseline and help ease future research in instance-level recognition. Code will be made available.
Видео Mask R-CNN канала ComputerVisionFoundation Videos
Показать
Комментарии отсутствуют
Информация о видео
17 ноября 2017 г. 3:45:51
00:12:22
Другие видео канала
![Lecture 11 | Detection and Segmentation](https://i.ytimg.com/vi/nDPWywWRIRo/default.jpg)
![The Code That Runs Our Lives](https://i.ytimg.com/vi/XG-dwZMc7Ng/default.jpg)
![YOLOv2 vs YOLOv3 vs Mask RCNN vs Deeplab Xception](https://i.ytimg.com/vi/s8Ui_kV9dhw/default.jpg)
![Mask RCNN with Keras and Tensorflow (pt.1) Setup and Installation](https://i.ytimg.com/vi/2TikTv6PWDw/default.jpg)
![Variational Autoencoders](https://i.ytimg.com/vi/9zKuYvjFFS8/default.jpg)
![Mask RCNN - COCO - instance segmentation](https://i.ytimg.com/vi/OOT3UIXZztE/default.jpg)
![Understanding Artificial Intelligence and Its Future | Neil Nie | TEDxDeerfield](https://i.ytimg.com/vi/SN2BZswEWUA/default.jpg)
![Advanced Image Recognition with Mask R-CNN, Aniket Rangrej,Ravi Ilango 20180919](https://i.ytimg.com/vi/9J5V6nbFtGI/default.jpg)
![Deep Residual Learning for Image Recognition](https://i.ytimg.com/vi/C6tLw-rPQ2o/default.jpg)
![](https://i.ytimg.com/vi/mdPxeprqwRY/default.jpg)
![Mask RCNN - How it Works - Intuition Tutorial](https://i.ytimg.com/vi/0vt05rQqk_I/default.jpg)
![Tutorial: Deep Learning for Objects and Scenes - Part 1](https://i.ytimg.com/vi/jHv37mKAhV4/default.jpg)
![Mask Region based Convolution Neural Networks - EXPLAINED!](https://i.ytimg.com/vi/4tkgOzQ9yyo/default.jpg)
![2017 Feature Pyramid Network for Object Detection (FPN) paper summary](https://i.ytimg.com/vi/mwMopcSRx1U/default.jpg)
![Generative Adversarial Networks (GANs) - Computerphile](https://i.ytimg.com/vi/Sw9r8CL98N0/default.jpg)
![MeshCNN: A Network with an Edge @ TWiML Online Meetup EMEA](https://i.ytimg.com/vi/s2yXMtt-6qo/default.jpg)
![MeshCNN: A Network with an Edge](https://i.ytimg.com/vi/EfrtgGDEoGI/default.jpg)
![Mask R-CNN Demo on Google Colab](https://i.ytimg.com/vi/l39TyxYH75E/default.jpg)
![Lecture 12 | Visualizing and Understanding](https://i.ytimg.com/vi/6wcs6szJWMY/default.jpg)
![Encoder Decoder Network - Computerphile](https://i.ytimg.com/vi/1icvxbAoPWc/default.jpg)