Pixellib

Keras model extracted from Ade20k tensorflow model's checkpoint.
source: https://github.com/tensorflow/models/blob/master/research/deeplab/g3doc/model_zoo.md

Mask R-CNN model trained on coco dataset.

This release contains deeplabv3+ models trained on pascalvoc dataset.
Tensorflow model.
source: https://github.com/tensorflow/models/blob/master/research/deeplab/g3doc/model_zoo.md

Keras model extracted from the tensorflow model's checkpoint.
source: https://github.com/bonlime/keras-deeplab-v3-plus

* PointRend is used for segmentation of objects in images and videos.
* Supports extraction of objects from their bounding boxes' coordinates and masks' values.
* Faster and more accurate than the tensorflow version:
*It achieves 0.26 seconds for processing a single image and 4fps for live camera feeds.*
*Using A TargetSize of 667 * 447: It achieves 0.20 seconds for processing a single image and 6fps for live camera feeds.*
*Using A TargetSize of 333 * 200: It achieves 0.15 seconds for processing a single image and 9fps for live camera feeds.*