Attend the Women in ML Symposium on December 7 Register now

Module: tfm.vision

Stay organized with collections Save and categorize content based on your preferences.

TensorFlow Models Vision Libraries.

Modules

anchor module: Anchor box and labeler definition.

anchor_generator module: Multi scale anchor generator definition.

augment module: Augmentation policies for enhanced image/video preprocessing.

backbones module: Backbones package definition.

box_matcher module: Box matcher implementation.

box_ops module: Box related ops.

classification_model module: Build classification models.

configs module: Configs package definition.

decoders module: Decoders package definition.

factory module: Factory methods to build models.

factory_3d module: Factory methods to build models.

heads module: Heads package definition.

image_classification module: Image classification task definition.

iou_similarity module: Region Similarity Calculators.

layers module: Layers package definition.

mask_ops module: Utility functions for segmentations.

maskrcnn module: MaskRCNN task definition.

maskrcnn_model module: R-CNN(-RS) models.

models module: Models under Vision package.

nms module: Tensorflow implementation of non max suppression.

preprocess_ops module: Preprocessing ops.

preprocess_ops_3d module: Utils for processing video dataset features.

retinanet module: RetinaNet task definition.

retinanet_model module: RetinaNet.

sampling_ops module: Class to subsample minibatches by balancing positives and negatives.

segmentation_model module: Build segmentation models.

semantic_segmentation module: Image segmentation task definition.

serving module: Tools for exporting models.

spatial_transform_ops module: Spatial transform ops.

target_gather module: Definition of target gather, which gathers targets from indices.

video_classification module: Video classification task definition.

video_classification_model module: Build video classification models.

Classes

class ImageClassificationTask: A task for image classification.

class MaskRCNNTask: A single-replica view of training procedure.

class RetinaNetTask: A single-replica view of training procedure.

class SemanticSegmentationTask: A task for semantic segmentation.

class VideoClassificationTask: A task for video classification.