Runs inference on a single image. Args: image: A PIL.Image object, raw input image. Returns: seg_map: Segmenta

Convenience function for accesssing box coordinates. Returns: a numpy array of shape [N, 4] representing box corners

Add data to a specified field. Args: field: a string parameter used to speficy a related field to be accessed. field_data: a numpy ar

Accesses data associated with the specified field in the box collection. Args: field: a string parameter used to speficy a related field to

Adds groundtruth for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. grou

Writes a config object to disk.

Labels boxes detected with the same class from the same image as tp/fp. Args: detected_boxes: A numpy array of shape [N, 4] representing de

Helper function for creating a Xception block. Args: scope: The scope of the block. depth_list: The depth of the bottleneck layer for each

Constructs the graph, creates a test session and returns the results. The graph is executed either on TPU or CPU based on the `tpu_test` flag.

Create test input tensor.

Return number of boxes held in collections.

Create test input tensor.

Helper function for creating a resnet_v1 beta variant bottleneck block. Args: scope: The scope of the block. base_depth: The depth of the b

Adds detections for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. detec

Return all non-box fields.

Convenience function for accessing masks. Returns: a numpy array of shape [N, height, width] representing masks

A shallow and thin ResNet v1 for faster tests.

A shallow and thin Xception for faster tests.

The main function that segments an entire sentence that contains Chinese characters into seperated words. Parameter:

Compute evaluation result. Args: relationships: A dictionary of numerical label-text label mapping; if specified, returns per-relat

Scales the input dimension. Args: dim: Input dimension (a scalar or a scalar Tensor). scale: The amount of scaling applied to the input.

Clears the state to prepare for a fresh evaluation.

Constructor. Args: categories: A list of dicts, each of which has the following keys - 'id': (required) an integer id uniquely iden

Gather boxes from BoxList according to indices and return new BoxList. By default, gather returns boxes corresponding to the input index list, as

This function creates an image that can be used to test vis functions. It makes an image composed of four colored rectangles. Returns:

Reads the image dimensions. Args: image_data: string of image data. Returns: image_height and image_width.

Returns a TF-Feature of bytes. Args: values: A string. Returns: A TF-Feature.

Classification task auc metrics. :param ground_truth: numpy matrix, (num_samples, 3) :param prediction: numpy matrix, (num_samples, 3)

Detection task auc metrics. :param ground_truth: numpy matrix, (num_samples, 1024, 512) :param prediction: numpy matrix, (num_samples, 10

Converts a BoxList containing 'masks' into a BoxMaskList. Args: boxlist: An np_box_list.BoxList object. Returns: An np_box_mask_list.Box

Gather boxes from np_box_mask_list.BoxMaskList according to indices. By default, gather returns boxes corresponding to the input index list, as w

Generator for v1 ResNet models (beta variant). This function generates a family of modified ResNet v1 models. In particular, the first original 7

Overlay labeled boxes on an image with formatted scores and label names. This function groups boxes that correspond to the same location and crea

Asserts successful updating of all learning rate schemes.

Compute CorLoc score for object detection result. Args: detected_boxes: A float numpy array of shape [N, 4], representing N regio

Returns a TF-Feature of int64_list. Args: values: A scalar or list of values. Returns: A TF-Feature.

Returns a boolean indicating whether the input is a tensor. Args: t: the input to be tested. Returns: a boolean that indicates whether t

Sets the 0-th dimension of the input tensor. Args: t: the input tensor, assuming the rank is at least 1. d0: an integer indicating the 0-th

Adds groundtruth for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. grou

Adds groundtruth for a single image to be used for evaluation. Args: image_id: A unique string/integer identifier for the image. grou

Computes area of boxes. Args: boxes: Numpy array with shape [N, 4] holding N boxes Returns: a numpy array with shape [N*1] representing

Computes area of masks. Args: masks: Numpy array with shape [N, height, width] holding N masks. Masks values are of type np.uint8 and val

Clears the state to prepare for a fresh evaluation.

Get corner coordinates of boxes. Returns: a list of 4 1-d numpy arrays [y_min, x_min, y_max, x_max]

Gets the network. Args: network_name: Network name. preprocess_images: Preprocesses the images or not. arg_scope: Optional, arg_scope t

Gets the logits for multi-scale inputs. The returned logits are all downsampled (due to max-pooling layers) for both training and evaluation.

:param serialized_example: :param output: (anything, 'shape', 'labels') :return:

Retains groundtruth by valid indices. Args: tensor_dict: a dictionary of following groundtruth tensors - fields.InputDataFields.groundtru

Splits a separable conv2d into depthwise and pointwise conv2d. This operation differs from `tf.layers.separable_conv2d` as this operation applies