| import numpy as np |
| import cv2 |
| import random |
|
|
|
|
| def get_training_data(dataset): |
| """ Reads in the predicted pos/rotation data as well as the ground truth """ |
| print(dataset) |
| images_batch = [] |
| timestamps = [] |
| po1 = [] |
| po2 = [] |
| with open(dataset) as f: |
| for line in f: |
| fname, p0, p1, p2, p3, p4, p5, p6 = line.split() |
| timestamps.append(fname) |
| img = np.load("path/to/txt_files/with/data/" + fname) |
| images_batch.append(img) |
| po1.append(np.array((np.float(p0), np.float(p1), np.float(p2)))) |
| po2.append(np.array((np.float(p3), np.float(p4), np.float(p5), np.float(p6)))) |
| return (images_batch, [np.asarray(po1), np.asarray(po2)]), timestamps |
|
|
|
|
| def limited_gen_data(source): |
| """ Yields the input/ground truth data in batches of size 1. Can be used for evaluating the test dataset""" |
| indices = range(len(source[0])) |
| for i in indices: |
| image = source[0][i] |
| pose_x_rel = source[1][0][i] |
| pose_q_rel = source[1][1][i] |
| yield image, pose_x_rel, pose_q_rel |
|
|
|
|
| def gen_data(source): |
| """ Data generator that randomly shuffles the training data and yields them as single elements. """ |
| while True: |
| indices = list(range(len(source[0]))) |
| random.shuffle(indices) |
| for i in indices: |
| image = source[0][i] |
| pose_x_rel = source[1][0][i] |
| pose_q_rel = source[1][1][i] |
| yield image, pose_x_rel, pose_q_rel |
| |
|
|
| def gen_data_batch(source, batchSize, im_shape): |
| """ Uses the gen_data and yields batches of the size batchSize. """ |
| data_gen = gen_data(source) |
| while True: |
| image_batch = [] |
| pose_x_batch = [] |
| pose_q_batch = [] |
| for _ in range(batchSize): |
| image, pose_x, pose_q = next(data_gen) |
| if image.shape[0] != im_shape[1] or image.shape[1] != im_shape[2]: |
| image = cv2.resize(image, (im_shape[2], im_shape[1])) |
| image_batch.append(image) |
| pose_x_batch.append(pose_x) |
| pose_q_batch.append(pose_q) |
| yield np.asarray(image_batch), [np.asarray(pose_x_batch), np.asarray(pose_q_batch)] |
|
|
