Module: tfm.vision.configs.maskrcnn

R-CNN(-RS) configuration definition.

Classes

class Anchor

class DataConfig: Input config for training.

class DetectionGenerator

class DetectionHead

class Losses

class MaskHead

class MaskRCNN

class MaskRCNNTask

class MaskROIAligner

class MaskSampler

class Parser

class ROIAligner

class ROIGenerator

class ROISampler

class RPNHead

Functions

cascadercnn_spinenet_coco(...): COCO object detection with Cascade RCNN-RS with SpineNet backbone.

fasterrcnn_resnetfpn_coco(...): COCO object detection with Faster R-CNN.

maskrcnn_resnetfpn_coco(...): COCO object detection with Mask R-CNN.

maskrcnn_spinenet_coco(...): COCO object detection with Mask R-CNN with SpineNet backbone.

COCO_INPUT_PATH_BASE 'coco'