tlc.integration.super_gradients.datasets.pose_dataset

Pose Estimation Dataset for SuperGradients.

Module Contents

Classes

Class	Description
PoseEstimationDataset	Dataset class for training pose estimation models on Animal Pose dataset.

Functions

Function	Description
hex_to_rgb	Convert #RRGGBB to (R, G, B)
rgb_to_hex	Convert (R, G, B) to #RRGGBB

API

class PoseEstimationDataset(

table: Table,

transforms: list[super_gradients.training.transforms.keypoint_transforms.AbstractKeypointTransform],

image_column: str = IMAGE,

keypoints_column: str = KEYPOINTS_2D,

)

Bases: super_gradients.training.datasets.pose_estimation_datasets.abstract_pose_estimation_dataset.AbstractPoseEstimationDataset

Dataset class for training pose estimation models on Animal Pose dataset.

check_table(

table: Table,

image_column: str = IMAGE,

keypoints_column: str = KEYPOINTS_2D,

) → tuple[int, list[tuple[int, int, int]] | None, list[tuple[int, int, int]] | None, list[tuple[int, int, int]] | None]

Check compatibility of the table with the dataset, and return keypoint and edge information

load_sample(

index: int,

) → super_gradients.training.samples.PoseEstimationSample

Load a sample from the dataset

hex_to_rgb(

hex_color: str,

) → tuple[int, int, int]

Convert #RRGGBB to (R, G, B)

rgb_to_hex(

rgb_color: tuple[int, int, int],

) → str

Convert (R, G, B) to #RRGGBB