| import rnn |
| import tensorflow as tf |
|
|
| from subnet_tf_utils import generative_cnn_encoder, generative_cnn_encoder_deeper, generative_cnn_encoder_deeper13, \ |
| generative_cnn_c3_encoder, generative_cnn_c3_encoder_deeper, generative_cnn_c3_encoder_deeper13, \ |
| generative_cnn_c3_encoder_combine33, generative_cnn_c3_encoder_combine43, \ |
| generative_cnn_c3_encoder_combine53, generative_cnn_c3_encoder_combineFC, \ |
| generative_cnn_c3_encoder_deeper13_attn |
| from rasterization_utils.NeuralRenderer import NeuralRasterizorStep |
| from vgg_utils.VGG16 import vgg_net_slim |
|
|
|
|
| class VirtualSketchingModel(object): |
| def __init__(self, hps, gpu_mode=True, reuse=False): |
| """Initializer for the model. |
| |
| Args: |
| hps: a HParams object containing model hyperparameters |
| gpu_mode: a boolean that when True, uses GPU mode. |
| reuse: a boolean that when true, attemps to reuse variables. |
| """ |
| self.hps = hps |
| assert hps.model_mode in ['train', 'eval', 'eval_sample', 'sample'] |
| |
| if not gpu_mode: |
| with tf.device('/cpu:0'): |
| print('Model using cpu.') |
| self.build_model() |
| else: |
| print('-' * 100) |
| print('model_mode:', hps.model_mode) |
| print('Model using gpu.') |
| self.build_model() |
|
|
| def build_model(self): |
| """Define model architecture.""" |
| self.config_model() |
|
|
| initial_state = self.get_decoder_inputs() |
| self.initial_state = initial_state |
| self.initial_state_list = tf.split(self.initial_state, self.total_loop, axis=0) |
|
|
| total_loss_list = [] |
| ras_loss_list = [] |
| perc_relu_raw_list = [] |
| perc_relu_norm_list = [] |
| sn_loss_list = [] |
| cursor_outside_loss_list = [] |
| win_size_outside_loss_list = [] |
| early_state_loss_list = [] |
|
|
| tower_grads = [] |
|
|
| pred_raster_imgs_list = [] |
| pred_raster_imgs_rgb_list = [] |
|
|
| for t_i in range(self.total_loop): |
| gpu_idx = t_i // self.hps.loop_per_gpu |
| gpu_i = self.hps.gpus[gpu_idx] |
| print(self.hps.model_mode, 'model, gpu:', gpu_i, ', loop:', t_i % self.hps.loop_per_gpu) |
| with tf.device('/gpu:%d' % gpu_i): |
| with tf.name_scope('GPU_%d' % gpu_i) as scope: |
| if t_i > 0: |
| tf.get_variable_scope().reuse_variables() |
| else: |
| total_loss_list.clear() |
| ras_loss_list.clear() |
| perc_relu_raw_list.clear() |
| perc_relu_norm_list.clear() |
| sn_loss_list.clear() |
| cursor_outside_loss_list.clear() |
| win_size_outside_loss_list.clear() |
| early_state_loss_list.clear() |
| tower_grads.clear() |
| pred_raster_imgs_list.clear() |
| pred_raster_imgs_rgb_list.clear() |
|
|
| split_input_photo = self.input_photo_list[t_i] |
| split_image_size = self.image_size[t_i] |
| split_init_cursor = self.init_cursor_list[t_i] |
| split_initial_state = self.initial_state_list[t_i] |
| if self.hps.input_channel == 1: |
| split_target_sketch = split_input_photo |
| else: |
| split_target_sketch = self.target_sketch_list[t_i] |
|
|
| |
| other_params, pen_ras, final_state, pred_raster_images, pred_raster_images_rgb, \ |
| pos_before_max_min, win_size_before_max_min \ |
| = self.get_points_and_raster_image(split_initial_state, split_init_cursor, split_input_photo, |
| split_image_size) |
| |
| |
| |
| |
|
|
| pred_raster_imgs = 1.0 - pred_raster_images |
| pred_raster_imgs_rgb = 1.0 - pred_raster_images_rgb |
| pred_raster_imgs_list.append(pred_raster_imgs) |
| pred_raster_imgs_rgb_list.append(pred_raster_imgs_rgb) |
|
|
| if not self.hps.use_softargmax: |
| pen_state_soft = pen_ras[:, 1:2] |
| else: |
| pen_state_soft = self.differentiable_argmax(pen_ras, self.hps.soft_beta) |
|
|
| pred_params = tf.concat([pen_state_soft, other_params], axis=1) |
| pred_params = tf.reshape(pred_params, shape=[-1, self.hps.max_seq_len, 7]) |
| |
|
|
| if self.hps.model_mode == 'train' or self.hps.model_mode == 'eval': |
| raster_cost, sn_cost, cursor_outside_cost, winsize_outside_cost, \ |
| early_pen_states_cost, \ |
| perc_relu_loss_raw, perc_relu_loss_norm = \ |
| self.build_losses(split_target_sketch, pred_raster_imgs, pred_params, |
| pos_before_max_min, win_size_before_max_min, |
| split_image_size) |
| |
|
|
| ras_loss_list.append(raster_cost) |
| perc_relu_raw_list.append(perc_relu_loss_raw) |
| perc_relu_norm_list.append(perc_relu_loss_norm) |
| sn_loss_list.append(sn_cost) |
| cursor_outside_loss_list.append(cursor_outside_cost) |
| win_size_outside_loss_list.append(winsize_outside_cost) |
| early_state_loss_list.append(early_pen_states_cost) |
|
|
| if self.hps.model_mode == 'train': |
| total_cost_split, grads_and_vars_split = self.build_training_op_split( |
| raster_cost, sn_cost, cursor_outside_cost, winsize_outside_cost, |
| early_pen_states_cost) |
| total_loss_list.append(total_cost_split) |
| tower_grads.append(grads_and_vars_split) |
|
|
| self.raster_cost = tf.reduce_mean(tf.stack(ras_loss_list, axis=0)) |
| self.perc_relu_losses_raw = tf.reduce_mean(tf.stack(perc_relu_raw_list, axis=0), axis=0) |
| self.perc_relu_losses_norm = tf.reduce_mean(tf.stack(perc_relu_norm_list, axis=0), axis=0) |
| self.stroke_num_cost = tf.reduce_mean(tf.stack(sn_loss_list, axis=0)) |
| self.pos_outside_cost = tf.reduce_mean(tf.stack(cursor_outside_loss_list, axis=0)) |
| self.win_size_outside_cost = tf.reduce_mean(tf.stack(win_size_outside_loss_list, axis=0)) |
| self.early_pen_states_cost = tf.reduce_mean(tf.stack(early_state_loss_list, axis=0)) |
| self.cost = tf.reduce_mean(tf.stack(total_loss_list, axis=0)) |
|
|
| self.pred_raster_imgs = tf.concat(pred_raster_imgs_list, axis=0) |
| self.pred_raster_imgs_rgb = tf.concat(pred_raster_imgs_rgb_list, axis=0) |
|
|
| if self.hps.model_mode == 'train': |
| self.build_training_op(tower_grads) |
|
|
| def config_model(self): |
| if self.hps.model_mode == 'train': |
| self.global_step = tf.Variable(0, name='global_step', trainable=False) |
|
|
| if self.hps.dec_model == 'lstm': |
| dec_cell_fn = rnn.LSTMCell |
| elif self.hps.dec_model == 'layer_norm': |
| dec_cell_fn = rnn.LayerNormLSTMCell |
| elif self.hps.dec_model == 'hyper': |
| dec_cell_fn = rnn.HyperLSTMCell |
| else: |
| assert False, 'please choose a respectable cell' |
|
|
| use_recurrent_dropout = self.hps.use_recurrent_dropout |
| use_input_dropout = self.hps.use_input_dropout |
| use_output_dropout = self.hps.use_output_dropout |
|
|
| dec_cell = dec_cell_fn( |
| self.hps.dec_rnn_size, |
| use_recurrent_dropout=use_recurrent_dropout, |
| dropout_keep_prob=self.hps.recurrent_dropout_prob) |
|
|
| |
| |
| |
| |
| if use_input_dropout: |
| print('Dropout to input w/ keep_prob = %4.4f.' % self.hps.input_dropout_prob) |
| dec_cell = tf.contrib.rnn.DropoutWrapper( |
| dec_cell, input_keep_prob=self.hps.input_dropout_prob) |
| if use_output_dropout: |
| print('Dropout to output w/ keep_prob = %4.4f.' % self.hps.output_dropout_prob) |
| dec_cell = tf.contrib.rnn.DropoutWrapper( |
| dec_cell, output_keep_prob=self.hps.output_dropout_prob) |
| self.dec_cell = dec_cell |
|
|
| self.total_loop = len(self.hps.gpus) * self.hps.loop_per_gpu |
|
|
| self.init_cursor = tf.placeholder( |
| dtype=tf.float32, |
| shape=[self.hps.batch_size, 1, 2]) |
| self.init_width = tf.placeholder( |
| dtype=tf.float32, |
| shape=[1]) |
| self.image_size = tf.placeholder(dtype=tf.int32, shape=(self.total_loop)) |
|
|
| self.init_cursor_list = tf.split(self.init_cursor, self.total_loop, axis=0) |
| self.input_photo_list = [] |
| for loop_i in range(self.total_loop): |
| input_photo_i = tf.placeholder(dtype=tf.float32, shape=[None, None, None, self.hps.input_channel]) |
| self.input_photo_list.append(input_photo_i) |
|
|
| if self.hps.input_channel == 3: |
| self.target_sketch_list = [] |
| for loop_i in range(self.total_loop): |
| target_sketch_i = tf.placeholder(dtype=tf.float32, shape=[None, None, None, 1]) |
| self.target_sketch_list.append(target_sketch_i) |
|
|
| if self.hps.model_mode == 'train' or self.hps.model_mode == 'eval': |
| self.stroke_num_loss_weight = tf.Variable(0.0, trainable=False) |
| self.early_pen_loss_start_idx = tf.Variable(0, dtype=tf.int32, trainable=False) |
| self.early_pen_loss_end_idx = tf.Variable(0, dtype=tf.int32, trainable=False) |
|
|
| if self.hps.model_mode == 'train': |
| self.perc_loss_mean_list = [] |
| for loop_i in range(len(self.hps.perc_loss_layers)): |
| relu_loss_mean = tf.Variable(0.0, trainable=False) |
| self.perc_loss_mean_list.append(relu_loss_mean) |
| self.last_step_num = tf.Variable(0.0, trainable=False) |
|
|
| with tf.variable_scope('train_op', reuse=tf.AUTO_REUSE): |
| self.lr = tf.Variable(self.hps.learning_rate, trainable=False) |
| self.optimizer = tf.train.AdamOptimizer(self.lr) |
|
|
| |
|
|
| def normalize_image_m1to1(self, in_img_0to1): |
| norm_img_m1to1 = tf.multiply(in_img_0to1, 2.0) |
| norm_img_m1to1 = tf.subtract(norm_img_m1to1, 1.0) |
| return norm_img_m1to1 |
|
|
| def add_coords(self, input_tensor): |
| batch_size_tensor = tf.shape(input_tensor)[0] |
|
|
| xx_ones = tf.ones([batch_size_tensor, self.hps.raster_size], dtype=tf.int32) |
| xx_ones = tf.expand_dims(xx_ones, -1) |
| xx_range = tf.tile(tf.expand_dims(tf.range(self.hps.raster_size), 0), |
| [batch_size_tensor, 1]) |
| xx_range = tf.expand_dims(xx_range, 1) |
|
|
| xx_channel = tf.matmul(xx_ones, xx_range) |
| xx_channel = tf.expand_dims(xx_channel, -1) |
|
|
| yy_ones = tf.ones([batch_size_tensor, self.hps.raster_size], dtype=tf.int32) |
| yy_ones = tf.expand_dims(yy_ones, 1) |
| yy_range = tf.tile(tf.expand_dims(tf.range(self.hps.raster_size), 0), |
| [batch_size_tensor, 1]) |
| yy_range = tf.expand_dims(yy_range, -1) |
|
|
| yy_channel = tf.matmul(yy_range, yy_ones) |
| yy_channel = tf.expand_dims(yy_channel, -1) |
|
|
| xx_channel = tf.cast(xx_channel, 'float32') / (self.hps.raster_size - 1) |
| yy_channel = tf.cast(yy_channel, 'float32') / (self.hps.raster_size - 1) |
| |
| |
|
|
| ret = tf.concat([ |
| input_tensor, |
| xx_channel, |
| yy_channel, |
| ], axis=-1) |
|
|
| return ret |
|
|
| def build_combined_encoder(self, patch_canvas, patch_photo, entire_canvas, entire_photo, cursor_pos, |
| image_size, window_size): |
| """ |
| :param patch_canvas: (N, raster_size, raster_size, 1), [-1.0-stroke, 1.0-BG] |
| :param patch_photo: (N, raster_size, raster_size, 1/3), [-1.0-stroke, 1.0-BG] |
| :param entire_canvas: (N, image_size, image_size, 1), [0.0-stroke, 1.0-BG] |
| :param entire_photo: (N, image_size, image_size, 1/3), [0.0-stroke, 1.0-BG] |
| :param cursor_pos: (N, 1, 2), in size [0.0, 1.0) |
| :param window_size: (N, 1, 1), float, in large size |
| :return: |
| """ |
| if self.hps.resize_method == 'BILINEAR': |
| resize_method = tf.image.ResizeMethod.BILINEAR |
| elif self.hps.resize_method == 'NEAREST_NEIGHBOR': |
| resize_method = tf.image.ResizeMethod.NEAREST_NEIGHBOR |
| elif self.hps.resize_method == 'BICUBIC': |
| resize_method = tf.image.ResizeMethod.BICUBIC |
| elif self.hps.resize_method == 'AREA': |
| resize_method = tf.image.ResizeMethod.AREA |
| else: |
| raise Exception('unknown resize_method', self.hps.resize_method) |
|
|
| patch_photo = tf.stop_gradient(patch_photo) |
| patch_canvas = tf.stop_gradient(patch_canvas) |
| cursor_pos = tf.stop_gradient(cursor_pos) |
| window_size = tf.stop_gradient(window_size) |
|
|
| entire_photo_small = tf.stop_gradient(tf.image.resize_images(entire_photo, |
| (self.hps.raster_size, self.hps.raster_size), |
| method=resize_method)) |
| entire_canvas_small = tf.stop_gradient(tf.image.resize_images(entire_canvas, |
| (self.hps.raster_size, self.hps.raster_size), |
| method=resize_method)) |
| entire_photo_small = self.normalize_image_m1to1(entire_photo_small) |
| entire_canvas_small = self.normalize_image_m1to1(entire_canvas_small) |
|
|
| if self.hps.encode_cursor_type == 'value': |
| cursor_pos_norm = tf.expand_dims(cursor_pos, axis=1) |
| cursor_pos_norm = tf.tile(cursor_pos_norm, [1, self.hps.raster_size, self.hps.raster_size, 1]) |
| cursor_info = cursor_pos_norm |
| else: |
| raise Exception('Unknown encode_cursor_type', self.hps.encode_cursor_type) |
|
|
| batch_input_combined = tf.concat([patch_photo, patch_canvas, entire_photo_small, entire_canvas_small, cursor_info], |
| axis=-1) |
| batch_input_local = tf.concat([patch_photo, patch_canvas], axis=-1) |
| batch_input_global = tf.concat([entire_photo_small, entire_canvas_small, cursor_info], |
| axis=-1) |
|
|
| if self.hps.model_mode == 'train': |
| is_training = True |
| dropout_keep_prob = self.hps.pix_drop_kp |
| else: |
| is_training = False |
| dropout_keep_prob = 1.0 |
|
|
| if self.hps.add_coordconv: |
| batch_input_combined = self.add_coords(batch_input_combined) |
| batch_input_local = self.add_coords(batch_input_local) |
| batch_input_global = self.add_coords(batch_input_global) |
|
|
| if 'combine' in self.hps.encoder_type: |
| if self.hps.encoder_type == 'combine33': |
| image_embedding, _ = generative_cnn_c3_encoder_combine33(batch_input_local, batch_input_global, |
| is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'combine43': |
| image_embedding, _ = generative_cnn_c3_encoder_combine43(batch_input_local, batch_input_global, |
| is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'combine53': |
| image_embedding, _ = generative_cnn_c3_encoder_combine53(batch_input_local, batch_input_global, |
| is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'combineFC': |
| image_embedding, _ = generative_cnn_c3_encoder_combineFC(batch_input_local, batch_input_global, |
| is_training, dropout_keep_prob) |
| else: |
| raise Exception('Unknown encoder_type', self.hps.encoder_type) |
| else: |
| with tf.variable_scope('Combined_Encoder', reuse=tf.AUTO_REUSE): |
| if self.hps.encoder_type == 'conv10': |
| image_embedding, _ = generative_cnn_encoder(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv10_deep': |
| image_embedding, _ = generative_cnn_encoder_deeper(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv13': |
| image_embedding, _ = generative_cnn_encoder_deeper13(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv10_c3': |
| image_embedding, _ = generative_cnn_c3_encoder(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv10_deep_c3': |
| image_embedding, _ = generative_cnn_c3_encoder_deeper(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv13_c3': |
| image_embedding, _ = generative_cnn_c3_encoder_deeper13(batch_input_combined, is_training, dropout_keep_prob) |
| elif self.hps.encoder_type == 'conv13_c3_attn': |
| image_embedding, _ = generative_cnn_c3_encoder_deeper13_attn(batch_input_combined, is_training, dropout_keep_prob) |
| else: |
| raise Exception('Unknown encoder_type', self.hps.encoder_type) |
| return image_embedding |
|
|
| def build_seq_decoder(self, dec_cell, actual_input_x, initial_state): |
| rnn_output, last_state = self.rnn_decoder(dec_cell, initial_state, actual_input_x) |
| rnn_output_flat = tf.reshape(rnn_output, [-1, self.hps.dec_rnn_size]) |
|
|
| pen_n_out = 2 |
| params_n_out = 6 |
|
|
| with tf.variable_scope('DEC_RNN_out_pen', reuse=tf.AUTO_REUSE): |
| output_w_pen = tf.get_variable('output_w', [self.hps.dec_rnn_size, pen_n_out]) |
| output_b_pen = tf.get_variable('output_b', [pen_n_out], initializer=tf.constant_initializer(0.0)) |
| output_pen = tf.nn.xw_plus_b(rnn_output_flat, output_w_pen, output_b_pen) |
|
|
| with tf.variable_scope('DEC_RNN_out_params', reuse=tf.AUTO_REUSE): |
| output_w_params = tf.get_variable('output_w', [self.hps.dec_rnn_size, params_n_out]) |
| output_b_params = tf.get_variable('output_b', [params_n_out], initializer=tf.constant_initializer(0.0)) |
| output_params = tf.nn.xw_plus_b(rnn_output_flat, output_w_params, output_b_params) |
|
|
| output = tf.concat([output_pen, output_params], axis=1) |
|
|
| return output, last_state |
|
|
| def get_mixture_coef(self, outputs): |
| z = outputs |
| z_pen_logits = z[:, 0:2] |
| z_other_params_logits = z[:, 2:] |
|
|
| z_pen = tf.nn.softmax(z_pen_logits) |
| if self.hps.position_format == 'abs': |
| x1y1 = tf.nn.sigmoid(z_other_params_logits[:, 0:2]) |
| x2y2 = tf.tanh(z_other_params_logits[:, 2:4]) |
| widths = tf.nn.sigmoid(z_other_params_logits[:, 4:5]) |
| widths = tf.add(tf.multiply(widths, 1.0 - self.hps.min_width), self.hps.min_width) |
| scaling = tf.nn.sigmoid(z_other_params_logits[:, 5:6]) * self.hps.max_scaling |
| |
| |
| z_other_params = tf.concat([x1y1, x2y2, widths, scaling], axis=-1) |
| else: |
| raise Exception('Unknown position_format', self.hps.position_format) |
|
|
| r = [z_other_params, z_pen] |
| return r |
|
|
| |
|
|
| def get_decoder_inputs(self): |
| initial_state = self.dec_cell.zero_state(batch_size=self.hps.batch_size, dtype=tf.float32) |
| return initial_state |
|
|
| def rnn_decoder(self, dec_cell, initial_state, actual_input_x): |
| with tf.variable_scope("RNN_DEC", reuse=tf.AUTO_REUSE): |
| output, last_state = tf.nn.dynamic_rnn( |
| dec_cell, |
| actual_input_x, |
| initial_state=initial_state, |
| time_major=False, |
| swap_memory=True, |
| dtype=tf.float32) |
| return output, last_state |
|
|
| |
|
|
| def image_padding(self, ori_image, window_size, pad_value): |
| """ |
| Pad with (bg) |
| :param ori_image: |
| :return: |
| """ |
| paddings = [[0, 0], |
| [window_size // 2, window_size // 2], |
| [window_size // 2, window_size // 2], |
| [0, 0]] |
| pad_img = tf.pad(ori_image, paddings=paddings, mode='CONSTANT', constant_values=pad_value) |
| return pad_img |
|
|
| def image_cropping_fn(self, fn_inputs): |
| """ |
| crop the patch |
| :return: |
| """ |
| index_offset = self.hps.input_channel - 1 |
| input_image = fn_inputs[:, :, 0:2 + index_offset] |
| cursor_pos = fn_inputs[0, 0, 2 + index_offset:4 + index_offset] |
| image_size = fn_inputs[0, 0, 4 + index_offset] |
| window_size = tf.cast(fn_inputs[0, 0, 5 + index_offset], tf.int32) |
|
|
| input_img_reshape = tf.expand_dims(input_image, axis=0) |
| pad_img = self.image_padding(input_img_reshape, window_size, pad_value=0.0) |
|
|
| cursor_pos = tf.cast(tf.round(tf.multiply(cursor_pos, image_size)), dtype=tf.int32) |
| x0, x1 = cursor_pos[0], cursor_pos[0] + window_size |
| y0, y1 = cursor_pos[1], cursor_pos[1] + window_size |
| patch_image = pad_img[:, y0:y1, x0:x1, :] |
|
|
| |
| patch_image_scaled = tf.image.resize_images(patch_image, (self.hps.raster_size, self.hps.raster_size), |
| method=tf.image.ResizeMethod.AREA) |
| patch_image_scaled = tf.squeeze(patch_image_scaled, axis=0) |
| |
|
|
| return patch_image_scaled |
|
|
| def image_cropping(self, cursor_position, input_img, image_size, window_sizes): |
| """ |
| :param cursor_position: (N, 1, 2), float type, in size [0.0, 1.0) |
| :param input_img: (N, image_size, image_size, 2/4), [0.0-BG, 1.0-stroke] |
| :param window_sizes: (N, 1, 1), float32, with grad |
| """ |
| input_img_ = input_img |
| window_sizes_non_grad = tf.stop_gradient(tf.round(window_sizes)) |
|
|
| cursor_position_ = tf.reshape(cursor_position, (-1, 1, 1, 2)) |
| cursor_position_ = tf.tile(cursor_position_, [1, image_size, image_size, 1]) |
|
|
| image_size_ = tf.reshape(tf.cast(image_size, tf.float32), (1, 1, 1, 1)) |
| image_size_ = tf.tile(image_size_, [self.hps.batch_size // self.total_loop, image_size, image_size, 1]) |
|
|
| window_sizes_ = tf.reshape(window_sizes_non_grad, (-1, 1, 1, 1)) |
| window_sizes_ = tf.tile(window_sizes_, [1, image_size, image_size, 1]) |
|
|
| fn_inputs = tf.concat([input_img_, cursor_position_, image_size_, window_sizes_], |
| axis=-1) |
| curr_patch_imgs = tf.map_fn(self.image_cropping_fn, fn_inputs, parallel_iterations=32) |
| return curr_patch_imgs |
|
|
| def image_cropping_v3(self, cursor_position, input_img, image_size, window_sizes): |
| """ |
| :param cursor_position: (N, 1, 2), float type, in size [0.0, 1.0) |
| :param input_img: (N, image_size, image_size, k), [0.0-BG, 1.0-stroke] |
| :param window_sizes: (N, 1, 1), float32, with grad |
| """ |
| window_sizes_non_grad = tf.stop_gradient(window_sizes) |
|
|
| cursor_pos = tf.multiply(cursor_position, tf.cast(image_size, tf.float32)) |
| cursor_x, cursor_y = tf.split(cursor_pos, 2, axis=-1) |
|
|
| y1 = cursor_y - (window_sizes_non_grad - 1.0) / 2 |
| x1 = cursor_x - (window_sizes_non_grad - 1.0) / 2 |
| y2 = y1 + (window_sizes_non_grad - 1.0) |
| x2 = x1 + (window_sizes_non_grad - 1.0) |
| boxes = tf.concat([y1, x1, y2, x2], axis=-1) |
| boxes = tf.squeeze(boxes, axis=1) |
| boxes = boxes / tf.cast(image_size - 1, tf.float32) |
|
|
| box_ind = tf.ones_like(cursor_x)[:, 0, 0] |
| box_ind = tf.cast(box_ind, dtype=tf.int32) |
| box_ind = tf.cumsum(box_ind) - 1 |
|
|
| curr_patch_imgs = tf.image.crop_and_resize(input_img, boxes, box_ind, |
| crop_size=[self.hps.raster_size, self.hps.raster_size]) |
| |
| return curr_patch_imgs |
|
|
| def get_pixel_value(self, img, x, y): |
| """ |
| Utility function to get pixel value for coordinate vectors x and y from a 4D tensor image. |
| |
| Input |
| ----- |
| - img: tensor of shape (B, H, W, C) |
| - x: flattened tensor of shape (B, H', W') |
| - y: flattened tensor of shape (B, H', W') |
| |
| Returns |
| ------- |
| - output: tensor of shape (B, H', W', C) |
| """ |
| shape = tf.shape(x) |
| batch_size = shape[0] |
| height = shape[1] |
| width = shape[2] |
|
|
| batch_idx = tf.range(0, batch_size) |
| batch_idx = tf.reshape(batch_idx, (batch_size, 1, 1)) |
| b = tf.tile(batch_idx, (1, height, width)) |
|
|
| indices = tf.stack([b, y, x], 3) |
|
|
| return tf.gather_nd(img, indices) |
|
|
| def image_pasting_nondiff_single(self, fn_inputs): |
| patch_image = fn_inputs[:, :, 0:1] |
| cursor_pos = fn_inputs[0, 0, 1:3] |
| image_size = tf.cast(fn_inputs[0, 0, 3], tf.int32) |
| window_size = tf.cast(fn_inputs[0, 0, 4], tf.int32) |
|
|
| patch_image_scaled = tf.expand_dims(patch_image, axis=0) |
| patch_image_scaled = tf.image.resize_images(patch_image_scaled, (window_size, window_size), |
| method=tf.image.ResizeMethod.BILINEAR) |
| patch_image_scaled = tf.squeeze(patch_image_scaled, axis=0) |
| |
|
|
| cursor_pos = tf.cast(tf.round(cursor_pos), dtype=tf.int32) |
| cursor_x, cursor_y = cursor_pos[0], cursor_pos[1] |
|
|
| pad_up = cursor_y |
| pad_down = image_size - cursor_y |
| pad_left = cursor_x |
| pad_right = image_size - cursor_x |
|
|
| paddings = [[pad_up, pad_down], |
| [pad_left, pad_right], |
| [0, 0]] |
| pad_img = tf.pad(patch_image_scaled, paddings=paddings, mode='CONSTANT', |
| constant_values=0.0) |
|
|
| crop_start = window_size // 2 |
| pasted_image = pad_img[crop_start: crop_start + image_size, crop_start: crop_start + image_size, :] |
| return pasted_image |
|
|
| def image_pasting_diff_single(self, fn_inputs): |
| patch_canvas = fn_inputs[:, :, 0:1] |
| cursor_pos = fn_inputs[0, 0, 1:3] |
| image_size = tf.cast(fn_inputs[0, 0, 3], tf.int32) |
| window_size = tf.cast(fn_inputs[0, 0, 4], tf.int32) |
| cursor_x, cursor_y = cursor_pos[0], cursor_pos[1] |
|
|
| patch_canvas_scaled = tf.expand_dims(patch_canvas, axis=0) |
| patch_canvas_scaled = tf.image.resize_images(patch_canvas_scaled, (window_size, window_size), |
| method=tf.image.ResizeMethod.BILINEAR) |
| |
|
|
| valid_canvas = self.image_pasting_diff_batch(patch_canvas_scaled, |
| tf.expand_dims(tf.expand_dims(cursor_pos, axis=0), axis=0), |
| window_size) |
| valid_canvas = tf.squeeze(valid_canvas, axis=0) |
| |
|
|
| pad_up = tf.cast(tf.floor(cursor_y), tf.int32) |
| pad_down = image_size - 1 - tf.cast(tf.floor(cursor_y), tf.int32) |
| pad_left = tf.cast(tf.floor(cursor_x), tf.int32) |
| pad_right = image_size - 1 - tf.cast(tf.floor(cursor_x), tf.int32) |
|
|
| paddings = [[pad_up, pad_down], |
| [pad_left, pad_right], |
| [0, 0]] |
| pad_img = tf.pad(valid_canvas, paddings=paddings, mode='CONSTANT', |
| constant_values=0.0) |
|
|
| crop_start = window_size // 2 |
| pasted_image = pad_img[crop_start: crop_start + image_size, crop_start: crop_start + image_size, :] |
| return pasted_image |
|
|
| def image_pasting_diff_single_v3(self, fn_inputs): |
| patch_canvas = fn_inputs[:, :, 0:1] |
| cursor_pos_a = fn_inputs[0, 0, 1:3] |
| image_size_a = tf.cast(fn_inputs[0, 0, 3], tf.int32) |
| window_size_a = fn_inputs[0, 0, 4] |
| raster_size_a = float(self.hps.raster_size) |
|
|
| padding_size = tf.cast(tf.ceil(window_size_a / 2.0), tf.int32) |
|
|
| x1y1_a = cursor_pos_a - window_size_a / 2.0 |
| x2y2_a = cursor_pos_a + window_size_a / 2.0 |
|
|
| x1y1_a_floor = tf.floor(x1y1_a) |
| x2y2_a_ceil = tf.ceil(x2y2_a) |
|
|
| cursor_pos_b_oricoord = (x1y1_a_floor + x2y2_a_ceil) / 2.0 |
| cursor_pos_b = (cursor_pos_b_oricoord - x1y1_a) / window_size_a * raster_size_a |
| raster_size_b = (x2y2_a_ceil - x1y1_a_floor) |
| image_size_b = raster_size_a |
| window_size_b = raster_size_a * (raster_size_b / window_size_a) |
|
|
| cursor_b_x, cursor_b_y = tf.split(cursor_pos_b, 2, axis=-1) |
|
|
| y1_b = cursor_b_y - (window_size_b[1] - 1.) / 2. |
| x1_b = cursor_b_x - (window_size_b[0] - 1.) / 2. |
| y2_b = y1_b + (window_size_b[1] - 1.) |
| x2_b = x1_b + (window_size_b[0] - 1.) |
| boxes_b = tf.concat([y1_b, x1_b, y2_b, x2_b], axis=-1) |
| boxes_b = boxes_b / tf.cast(image_size_b - 1, tf.float32) |
|
|
| box_ind_b = tf.ones((1), dtype=tf.int32) |
| box_ind_b = tf.cumsum(box_ind_b) - 1 |
|
|
| patch_canvas = tf.expand_dims(patch_canvas, axis=0) |
| boxes_b = tf.expand_dims(boxes_b, axis=0) |
|
|
| valid_canvas = tf.image.crop_and_resize(patch_canvas, boxes_b, box_ind_b, |
| crop_size=[raster_size_b[1], raster_size_b[0]]) |
| valid_canvas = valid_canvas[0] |
|
|
| pad_up = tf.cast(x1y1_a_floor[1], tf.int32) + padding_size |
| pad_down = image_size_a + padding_size - tf.cast(x2y2_a_ceil[1], tf.int32) |
| pad_left = tf.cast(x1y1_a_floor[0], tf.int32) + padding_size |
| pad_right = image_size_a + padding_size - tf.cast(x2y2_a_ceil[0], tf.int32) |
|
|
| paddings = [[pad_up, pad_down], |
| [pad_left, pad_right], |
| [0, 0]] |
| pad_img = tf.pad(valid_canvas, paddings=paddings, mode='CONSTANT', |
| constant_values=0.0) |
|
|
| pasted_image = pad_img[padding_size: padding_size + image_size_a, padding_size: padding_size + image_size_a, :] |
| return pasted_image |
|
|
| def image_pasting_diff_batch(self, patch_image, cursor_position, window_size): |
| """ |
| :param patch_img: (N, window_size, window_size, 1), [0.0-BG, 1.0-stroke] |
| :param cursor_position: (N, 1, 2), in large size |
| :return: |
| """ |
| paddings1 = [[0, 0], |
| [1, 1], |
| [1, 1], |
| [0, 0]] |
| patch_image_pad1 = tf.pad(patch_image, paddings=paddings1, mode='CONSTANT', |
| constant_values=0.0) |
|
|
| cursor_x, cursor_y = cursor_position[:, :, 0:1], cursor_position[:, :, 1:2] |
| cursor_x_f, cursor_y_f = tf.floor(cursor_x), tf.floor(cursor_y) |
| patch_x, patch_y = 1.0 - (cursor_x - cursor_x_f), 1.0 - (cursor_y - cursor_y_f) |
|
|
| x_ones = tf.ones_like(patch_x, dtype=tf.float32) |
| x_ones = tf.tile(x_ones, [1, 1, window_size]) |
| patch_x = tf.concat([patch_x, x_ones], axis=-1) |
| patch_x = tf.tile(patch_x, [1, window_size + 1, 1]) |
| patch_x = tf.cumsum(patch_x, axis=-1) |
| patch_x0 = tf.cast(tf.floor(patch_x), tf.int32) |
| patch_x1 = patch_x0 + 1 |
|
|
| y_ones = tf.ones_like(patch_y, dtype=tf.float32) |
| y_ones = tf.tile(y_ones, [1, window_size, 1]) |
| patch_y = tf.concat([patch_y, y_ones], axis=1) |
| patch_y = tf.tile(patch_y, [1, 1, window_size + 1]) |
| patch_y = tf.cumsum(patch_y, axis=1) |
| patch_y0 = tf.cast(tf.floor(patch_y), tf.int32) |
| patch_y1 = patch_y0 + 1 |
|
|
| |
| valid_canvas_patch_a = self.get_pixel_value(patch_image_pad1, patch_x0, patch_y0) |
| valid_canvas_patch_b = self.get_pixel_value(patch_image_pad1, patch_x0, patch_y1) |
| valid_canvas_patch_c = self.get_pixel_value(patch_image_pad1, patch_x1, patch_y0) |
| valid_canvas_patch_d = self.get_pixel_value(patch_image_pad1, patch_x1, patch_y1) |
| |
|
|
| patch_x0 = tf.cast(patch_x0, tf.float32) |
| patch_x1 = tf.cast(patch_x1, tf.float32) |
| patch_y0 = tf.cast(patch_y0, tf.float32) |
| patch_y1 = tf.cast(patch_y1, tf.float32) |
|
|
| |
| wa = (patch_x1 - patch_x) * (patch_y1 - patch_y) |
| wb = (patch_x1 - patch_x) * (patch_y - patch_y0) |
| wc = (patch_x - patch_x0) * (patch_y1 - patch_y) |
| wd = (patch_x - patch_x0) * (patch_y - patch_y0) |
| |
|
|
| |
| wa = tf.expand_dims(wa, axis=3) |
| wb = tf.expand_dims(wb, axis=3) |
| wc = tf.expand_dims(wc, axis=3) |
| wd = tf.expand_dims(wd, axis=3) |
| |
|
|
| |
| valid_canvas_patch_ = tf.add_n([wa * valid_canvas_patch_a, |
| wb * valid_canvas_patch_b, |
| wc * valid_canvas_patch_c, |
| wd * valid_canvas_patch_d]) |
| return valid_canvas_patch_ |
|
|
| def image_pasting(self, cursor_position_norm, patch_img, image_size, window_sizes, is_differentiable=False): |
| """ |
| paste the patch_img to padded size based on cursor_position |
| :param cursor_position_norm: (N, 1, 2), float type, in size [0.0, 1.0) |
| :param patch_img: (N, raster_size, raster_size), [0.0-BG, 1.0-stroke] |
| :param window_sizes: (N, 1, 1), float32, with grad |
| :return: |
| """ |
| cursor_position = tf.multiply(cursor_position_norm, tf.cast(image_size, tf.float32)) |
| window_sizes_r = tf.round(window_sizes) |
|
|
| patch_img_ = tf.expand_dims(patch_img, axis=-1) |
| cursor_position_step = tf.reshape(cursor_position, (-1, 1, 1, 2)) |
| cursor_position_step = tf.tile(cursor_position_step, [1, self.hps.raster_size, self.hps.raster_size, |
| 1]) |
| image_size_tile = tf.reshape(tf.cast(image_size, tf.float32), (1, 1, 1, 1)) |
| image_size_tile = tf.tile(image_size_tile, [self.hps.batch_size // self.total_loop, self.hps.raster_size, |
| self.hps.raster_size, 1]) |
| window_sizes_tile = tf.reshape(window_sizes_r, (-1, 1, 1, 1)) |
| window_sizes_tile = tf.tile(window_sizes_tile, [1, self.hps.raster_size, self.hps.raster_size, 1]) |
|
|
| pasting_inputs = tf.concat([patch_img_, cursor_position_step, image_size_tile, window_sizes_tile], |
| axis=-1) |
|
|
| if is_differentiable: |
| curr_paste_imgs = tf.map_fn(self.image_pasting_diff_single, pasting_inputs, |
| parallel_iterations=32) |
| else: |
| curr_paste_imgs = tf.map_fn(self.image_pasting_nondiff_single, pasting_inputs, |
| parallel_iterations=32) |
| curr_paste_imgs = tf.squeeze(curr_paste_imgs, axis=-1) |
| return curr_paste_imgs |
|
|
| def image_pasting_v3(self, cursor_position_norm, patch_img, image_size, window_sizes, is_differentiable=False): |
| """ |
| paste the patch_img to padded size based on cursor_position |
| :param cursor_position_norm: (N, 1, 2), float type, in size [0.0, 1.0) |
| :param patch_img: (N, raster_size, raster_size), [0.0-BG, 1.0-stroke] |
| :param window_sizes: (N, 1, 1), float32, with grad |
| :return: |
| """ |
| cursor_position = tf.multiply(cursor_position_norm, tf.cast(image_size, tf.float32)) |
|
|
| if is_differentiable: |
| patch_img_ = tf.expand_dims(patch_img, axis=-1) |
| cursor_position_step = tf.reshape(cursor_position, (-1, 1, 1, 2)) |
| cursor_position_step = tf.tile(cursor_position_step, [1, self.hps.raster_size, self.hps.raster_size, |
| 1]) |
| image_size_tile = tf.reshape(tf.cast(image_size, tf.float32), (1, 1, 1, 1)) |
| image_size_tile = tf.tile(image_size_tile, [self.hps.batch_size // self.total_loop, self.hps.raster_size, |
| self.hps.raster_size, 1]) |
| window_sizes_tile = tf.reshape(window_sizes, (-1, 1, 1, 1)) |
| window_sizes_tile = tf.tile(window_sizes_tile, [1, self.hps.raster_size, self.hps.raster_size, 1]) |
|
|
| pasting_inputs = tf.concat([patch_img_, cursor_position_step, image_size_tile, window_sizes_tile], |
| axis=-1) |
| curr_paste_imgs = tf.map_fn(self.image_pasting_diff_single_v3, pasting_inputs, |
| parallel_iterations=32) |
| else: |
| raise Exception('Unfinished...') |
| curr_paste_imgs = tf.squeeze(curr_paste_imgs, axis=-1) |
| return curr_paste_imgs |
|
|
| def get_points_and_raster_image(self, initial_state, init_cursor, input_photo, image_size): |
| |
| prev_state = initial_state |
|
|
| prev_width = self.init_width |
| prev_width = tf.expand_dims(tf.expand_dims(prev_width, axis=0), axis=0) |
| prev_width = tf.tile(prev_width, [self.hps.batch_size // self.total_loop, 1, 1]) |
|
|
| prev_scaling = tf.ones((self.hps.batch_size // self.total_loop, 1, 1)) |
| prev_window_size = tf.ones((self.hps.batch_size // self.total_loop, 1, 1), |
| dtype=tf.float32) * float(self.hps.raster_size) |
|
|
| cursor_position_temp = init_cursor |
| self.cursor_position = cursor_position_temp |
| cursor_position_loop = self.cursor_position |
|
|
| other_params_list = [] |
| pen_ras_list = [] |
|
|
| pos_before_max_min_list = [] |
| win_size_before_max_min_list = [] |
|
|
| curr_canvas_soft = tf.zeros_like(input_photo[:, :, :, 0]) |
| curr_canvas_soft_rgb = tf.tile(tf.zeros_like(input_photo[:, :, :, 0:1]), [1, 1, 1, 3]) |
| curr_canvas_hard = tf.zeros_like(curr_canvas_soft) |
|
|
| |
| self.curr_canvas_hard = curr_canvas_hard |
|
|
| rasterizor_st = NeuralRasterizorStep( |
| raster_size=self.hps.raster_size, |
| position_format=self.hps.position_format) |
|
|
| if self.hps.cropping_type == 'v3': |
| cropping_func = self.image_cropping_v3 |
| |
| |
| else: |
| raise Exception('Unknown cropping_type', self.hps.cropping_type) |
|
|
| if self.hps.pasting_type == 'v3': |
| pasting_func = self.image_pasting_v3 |
| |
| |
| else: |
| raise Exception('Unknown pasting_type', self.hps.pasting_type) |
|
|
| for time_i in range(self.hps.max_seq_len): |
| cursor_position_non_grad = tf.stop_gradient(cursor_position_loop) |
|
|
| curr_window_size = tf.multiply(prev_scaling, tf.stop_gradient(prev_window_size)) |
| curr_window_size = tf.maximum(curr_window_size, tf.cast(self.hps.min_window_size, tf.float32)) |
| curr_window_size = tf.minimum(curr_window_size, tf.cast(image_size, tf.float32)) |
|
|
| |
| |
| curr_canvas_hard_non_grad = tf.stop_gradient(self.curr_canvas_hard) |
| curr_canvas_hard_non_grad = tf.expand_dims(curr_canvas_hard_non_grad, axis=-1) |
|
|
| |
| crop_inputs = tf.concat([1.0 - input_photo, curr_canvas_hard_non_grad], axis=-1) |
|
|
| cropped_outputs = cropping_func(cursor_position_non_grad, crop_inputs, image_size, curr_window_size) |
| index_offset = self.hps.input_channel - 1 |
| curr_patch_inputs = cropped_outputs[:, :, :, 0:1 + index_offset] |
| curr_patch_canvas_hard_non_grad = cropped_outputs[:, :, :, 1 + index_offset:2 + index_offset] |
| |
|
|
| curr_patch_inputs = 1.0 - curr_patch_inputs |
| curr_patch_inputs = self.normalize_image_m1to1(curr_patch_inputs) |
| |
|
|
| |
| curr_patch_canvas_hard_non_grad = 1.0 - curr_patch_canvas_hard_non_grad |
| curr_patch_canvas_hard_non_grad = self.normalize_image_m1to1(curr_patch_canvas_hard_non_grad) |
|
|
| |
| combined_z = self.build_combined_encoder( |
| curr_patch_canvas_hard_non_grad, |
| curr_patch_inputs, |
| 1.0 - curr_canvas_hard_non_grad, |
| input_photo, |
| cursor_position_non_grad, |
| image_size, |
| curr_window_size) |
| combined_z = tf.expand_dims(combined_z, axis=1) |
|
|
| curr_window_size_top_side_norm_non_grad = \ |
| tf.stop_gradient(curr_window_size / tf.cast(image_size, tf.float32)) |
| curr_window_size_bottom_side_norm_non_grad = \ |
| tf.stop_gradient(curr_window_size / tf.cast(self.hps.min_window_size, tf.float32)) |
| if not self.hps.concat_win_size: |
| combined_z = tf.concat([tf.stop_gradient(prev_width), combined_z], 2) |
| else: |
| combined_z = tf.concat([tf.stop_gradient(prev_width), |
| curr_window_size_top_side_norm_non_grad, |
| curr_window_size_bottom_side_norm_non_grad, |
| combined_z], |
| 2) |
|
|
| if self.hps.concat_cursor: |
| prev_input_x = tf.concat([cursor_position_non_grad, combined_z], 2) |
| else: |
| prev_input_x = combined_z |
|
|
| h_output, next_state = self.build_seq_decoder(self.dec_cell, prev_input_x, prev_state) |
| |
| [o_other_params, o_pen_ras] = self.get_mixture_coef(h_output) |
| |
| |
|
|
| o_other_params = tf.reshape(o_other_params, [-1, 1, 6]) |
| o_pen_ras_raw = tf.reshape(o_pen_ras, [-1, 1, 2]) |
|
|
| other_params_list.append(o_other_params) |
| pen_ras_list.append(o_pen_ras_raw) |
|
|
| |
|
|
| if self.hps.model_mode == 'train' or self.hps.model_mode == 'eval' or self.hps.model_mode == 'eval_sample': |
| |
| curr_other_params = tf.squeeze(o_other_params, axis=1) |
| x1y1, x2y2, width2, scaling = curr_other_params[:, 0:2], curr_other_params[:, 2:4],\ |
| curr_other_params[:, 4:5], curr_other_params[:, 5:6] |
| x0y0 = tf.zeros_like(x2y2) |
| x0y0 = tf.div(tf.add(x0y0, 1.0), 2.0) |
| x2y2 = tf.div(tf.add(x2y2, 1.0), 2.0) |
| widths = tf.concat([tf.squeeze(prev_width, axis=1), width2], axis=1) |
| curr_other_params = tf.concat([x0y0, x1y1, x2y2, widths], axis=-1) |
| curr_stroke_image = rasterizor_st.raster_func_stroke_abs(curr_other_params) |
| |
|
|
| curr_stroke_image_large = pasting_func(cursor_position_loop, curr_stroke_image, |
| image_size, curr_window_size, |
| is_differentiable=self.hps.pasting_diff) |
| |
|
|
| |
| if not self.hps.use_softargmax: |
| curr_state_soft = o_pen_ras[:, 1:2] |
| else: |
| curr_state_soft = self.differentiable_argmax(o_pen_ras, self.hps.soft_beta) |
|
|
| curr_state_soft = tf.expand_dims(curr_state_soft, axis=1) |
|
|
| filter_curr_stroke_image_soft = tf.multiply(tf.subtract(1.0, curr_state_soft), curr_stroke_image_large) |
| |
| curr_canvas_soft = tf.add(curr_canvas_soft, filter_curr_stroke_image_soft) |
|
|
| |
| curr_state_hard = tf.expand_dims(tf.cast(tf.argmax(o_pen_ras_raw, axis=-1), dtype=tf.float32), |
| axis=-1) |
| filter_curr_stroke_image_hard = tf.multiply(tf.subtract(1.0, curr_state_hard), curr_stroke_image_large) |
| |
| self.curr_canvas_hard = tf.add(self.curr_canvas_hard, filter_curr_stroke_image_hard) |
| self.curr_canvas_hard = tf.clip_by_value(self.curr_canvas_hard, 0.0, 1.0) |
|
|
| next_width = o_other_params[:, :, 4:5] |
| next_scaling = o_other_params[:, :, 5:6] |
| next_window_size = tf.multiply(next_scaling, tf.stop_gradient(curr_window_size)) |
| window_size_before_max_min = next_window_size |
| win_size_before_max_min_list.append(window_size_before_max_min) |
| next_window_size = tf.maximum(next_window_size, tf.cast(self.hps.min_window_size, tf.float32)) |
| next_window_size = tf.minimum(next_window_size, tf.cast(image_size, tf.float32)) |
|
|
| prev_state = next_state |
| prev_width = next_width * curr_window_size / next_window_size |
| prev_scaling = next_scaling |
| prev_window_size = curr_window_size |
|
|
| |
| new_cursor_offsets = tf.multiply(o_other_params[:, :, 2:4], |
| tf.divide(curr_window_size, 2.0)) |
| new_cursor_offset_next = new_cursor_offsets |
| new_cursor_offset_next = tf.concat([new_cursor_offset_next[:, :, 1:2], new_cursor_offset_next[:, :, 0:1]], axis=-1) |
|
|
| cursor_position_loop_large = tf.multiply(cursor_position_loop, tf.cast(image_size, tf.float32)) |
|
|
| if self.hps.stop_accu_grad: |
| stroke_position_next = tf.stop_gradient(cursor_position_loop_large) + new_cursor_offset_next |
| else: |
| stroke_position_next = cursor_position_loop_large + new_cursor_offset_next |
|
|
| stroke_position_before_max_min = stroke_position_next |
| pos_before_max_min_list.append(stroke_position_before_max_min) |
|
|
| if self.hps.cursor_type == 'next': |
| cursor_position_loop_large = stroke_position_next |
| else: |
| raise Exception('Unknown cursor_type') |
|
|
| cursor_position_loop_large = tf.maximum(cursor_position_loop_large, 0.0) |
| cursor_position_loop_large = tf.minimum(cursor_position_loop_large, tf.cast(image_size - 1, tf.float32)) |
| cursor_position_loop = tf.div(cursor_position_loop_large, tf.cast(image_size, tf.float32)) |
|
|
| curr_canvas_soft = tf.clip_by_value(curr_canvas_soft, 0.0, 1.0) |
|
|
| other_params_ = tf.reshape(tf.concat(other_params_list, axis=1), [-1, 6]) |
| pen_ras_ = tf.reshape(tf.concat(pen_ras_list, axis=1), [-1, 2]) |
| pos_before_max_min_ = tf.concat(pos_before_max_min_list, axis=1) |
| win_size_before_max_min_ = tf.concat(win_size_before_max_min_list, axis=1) |
|
|
| return other_params_, pen_ras_, prev_state, curr_canvas_soft, curr_canvas_soft_rgb, \ |
| pos_before_max_min_, win_size_before_max_min_ |
|
|
| def differentiable_argmax(self, input_pen, soft_beta): |
| """ |
| Differentiable argmax trick. |
| :param input_pen: (N, n_class) |
| :return: pen_state: (N, 1) |
| """ |
| def sign_onehot(x): |
| """ |
| :param x: (N, n_class) |
| :return: (N, n_class) |
| """ |
| y = tf.sign(tf.reduce_max(x, axis=-1, keepdims=True) - x) |
| y = (y - 1) * (-1) |
| return y |
|
|
| def softargmax(x, beta=1e2): |
| """ |
| :param x: (N, n_class) |
| :param beta: 1e10 is the best. 1e2 is acceptable. |
| :return: (N) |
| """ |
| x_range = tf.cumsum(tf.ones_like(x), axis=1) |
| return tf.reduce_sum(tf.nn.softmax(x * beta) * x_range, axis=1) - 1 |
|
|
| |
| |
| |
| pen_state = softargmax(input_pen, soft_beta) |
| pen_state = tf.expand_dims(pen_state, axis=1) |
| return pen_state |
|
|
| def build_losses(self, target_sketch, pred_raster_imgs, pred_params, |
| pos_before_max_min, win_size_before_max_min, image_size): |
| def get_raster_loss(pred_imgs, gt_imgs, loss_type): |
| perc_layer_losses_raw = [] |
| perc_layer_losses_weighted = [] |
| perc_layer_losses_norm = [] |
|
|
| if loss_type == 'l1': |
| ras_cost = tf.reduce_mean(tf.abs(tf.subtract(gt_imgs, pred_imgs))) |
| elif loss_type == 'l1_small': |
| gt_imgs_small = tf.image.resize_images(tf.expand_dims(gt_imgs, axis=3), (32, 32)) |
| pred_imgs_small = tf.image.resize_images(tf.expand_dims(pred_imgs, axis=3), (32, 32)) |
| ras_cost = tf.reduce_mean(tf.abs(tf.subtract(gt_imgs_small, pred_imgs_small))) |
| elif loss_type == 'mse': |
| ras_cost = tf.reduce_mean(tf.pow(tf.subtract(gt_imgs, pred_imgs), 2)) |
| elif loss_type == 'perceptual': |
| return_map_pred = vgg_net_slim(pred_imgs, image_size) |
| return_map_gt = vgg_net_slim(gt_imgs, image_size) |
| perc_loss_type = 'l1' |
| weighted_map = {'ReLU1_1': 100.0, 'ReLU1_2': 100.0, |
| 'ReLU2_1': 100.0, 'ReLU2_2': 100.0, |
| 'ReLU3_1': 10.0, 'ReLU3_2': 10.0, 'ReLU3_3': 10.0, |
| 'ReLU4_1': 1.0, 'ReLU4_2': 1.0, 'ReLU4_3': 1.0, |
| 'ReLU5_1': 1.0, 'ReLU5_2': 1.0, 'ReLU5_3': 1.0} |
|
|
| for perc_layer in self.hps.perc_loss_layers: |
| if perc_loss_type == 'l1': |
| perc_layer_loss = tf.reduce_mean(tf.abs(tf.subtract(return_map_pred[perc_layer], |
| return_map_gt[perc_layer]))) |
| elif perc_loss_type == 'mse': |
| perc_layer_loss = tf.reduce_mean(tf.pow(tf.subtract(return_map_pred[perc_layer], |
| return_map_gt[perc_layer]), 2)) |
| else: |
| raise NameError('Unknown perceptual loss type:', perc_loss_type) |
| perc_layer_losses_raw.append(perc_layer_loss) |
|
|
| assert perc_layer in weighted_map |
| perc_layer_losses_weighted.append(perc_layer_loss * weighted_map[perc_layer]) |
|
|
| for loop_i in range(len(self.hps.perc_loss_layers)): |
| perc_relu_loss_raw = perc_layer_losses_raw[loop_i] |
|
|
| if self.hps.model_mode == 'train': |
| curr_relu_mean = (self.perc_loss_mean_list[loop_i] * self.last_step_num + perc_relu_loss_raw) / (self.last_step_num + 1.0) |
| relu_cost_norm = perc_relu_loss_raw / curr_relu_mean |
| else: |
| relu_cost_norm = perc_relu_loss_raw |
| perc_layer_losses_norm.append(relu_cost_norm) |
|
|
| perc_layer_losses_raw = tf.stack(perc_layer_losses_raw, axis=0) |
| perc_layer_losses_norm = tf.stack(perc_layer_losses_norm, axis=0) |
|
|
| if self.hps.perc_loss_fuse_type == 'max': |
| ras_cost = tf.reduce_max(perc_layer_losses_norm) |
| elif self.hps.perc_loss_fuse_type == 'add': |
| ras_cost = tf.reduce_mean(perc_layer_losses_norm) |
| elif self.hps.perc_loss_fuse_type == 'raw_add': |
| ras_cost = tf.reduce_mean(perc_layer_losses_raw) |
| elif self.hps.perc_loss_fuse_type == 'weighted_sum': |
| ras_cost = tf.reduce_mean(perc_layer_losses_weighted) |
| else: |
| raise NameError('Unknown perc_loss_fuse_type:', self.hps.perc_loss_fuse_type) |
|
|
| elif loss_type == 'triplet': |
| raise Exception('Solution for triplet loss is coming soon.') |
| else: |
| raise NameError('Unknown loss type:', loss_type) |
|
|
| if loss_type != 'perceptual': |
| for perc_layer_i in self.hps.perc_loss_layers: |
| perc_layer_losses_raw.append(tf.constant(0.0)) |
| perc_layer_losses_norm.append(tf.constant(0.0)) |
|
|
| perc_layer_losses_raw = tf.stack(perc_layer_losses_raw, axis=0) |
| perc_layer_losses_norm = tf.stack(perc_layer_losses_norm, axis=0) |
|
|
| return ras_cost, perc_layer_losses_raw, perc_layer_losses_norm |
|
|
| gt_raster_images = tf.squeeze(target_sketch, axis=3) |
| raster_cost, perc_relu_losses_raw, perc_relu_losses_norm = \ |
| get_raster_loss(pred_raster_imgs, gt_raster_images, loss_type=self.hps.raster_loss_base_type) |
|
|
| def get_stroke_num_loss(input_strokes): |
| ending_state = input_strokes[:, :, 0] |
| stroke_num_loss_pre = tf.reduce_mean(ending_state) |
| stroke_num_loss = 1.0 - stroke_num_loss_pre |
| return stroke_num_loss |
|
|
| stroke_num_cost = get_stroke_num_loss(pred_params) |
|
|
| def get_pos_outside_loss(pos_before_max_min_): |
| pos_after_max_min = tf.maximum(pos_before_max_min_, 0.0) |
| pos_after_max_min = tf.minimum(pos_after_max_min, tf.cast(image_size - 1, tf.float32)) |
| pos_outside_loss = tf.reduce_mean(tf.abs(pos_before_max_min_ - pos_after_max_min)) |
| return pos_outside_loss |
|
|
| pos_outside_cost = get_pos_outside_loss(pos_before_max_min) |
|
|
| def get_win_size_outside_loss(win_size_before_max_min_, min_window_size): |
| win_size_outside_top_loss = tf.divide( |
| tf.maximum(win_size_before_max_min_ - tf.cast(image_size, tf.float32), 0.0), |
| tf.cast(image_size, tf.float32)) |
| win_size_outside_bottom_loss = tf.divide( |
| tf.maximum(tf.cast(min_window_size, tf.float32) - win_size_before_max_min_, 0.0), |
| tf.cast(min_window_size, tf.float32)) |
| win_size_outside_loss = tf.reduce_mean(win_size_outside_top_loss + win_size_outside_bottom_loss) |
| return win_size_outside_loss |
|
|
| win_size_outside_cost = get_win_size_outside_loss(win_size_before_max_min, self.hps.min_window_size) |
|
|
| def get_early_pen_states_loss(input_strokes, curr_start, curr_end): |
| |
| pred_early_pen_states = input_strokes[:, curr_start:curr_end, 0] |
| pred_early_pen_states_min = tf.reduce_min(pred_early_pen_states, axis=1) |
| early_pen_states_loss = tf.reduce_mean(pred_early_pen_states_min) |
| return early_pen_states_loss |
|
|
| early_pen_states_cost = get_early_pen_states_loss(pred_params, |
| self.early_pen_loss_start_idx, self.early_pen_loss_end_idx) |
|
|
| return raster_cost, stroke_num_cost, pos_outside_cost, win_size_outside_cost, \ |
| early_pen_states_cost, \ |
| perc_relu_losses_raw, perc_relu_losses_norm |
|
|
| def build_training_op_split(self, raster_cost, sn_cost, cursor_outside_cost, win_size_outside_cost, |
| early_pen_states_cost): |
| total_cost = self.hps.raster_loss_weight * raster_cost + \ |
| self.hps.early_pen_loss_weight * early_pen_states_cost + \ |
| self.stroke_num_loss_weight * sn_cost + \ |
| self.hps.outside_loss_weight * cursor_outside_cost + \ |
| self.hps.win_size_outside_loss_weight * win_size_outside_cost |
|
|
| tvars = [var for var in tf.trainable_variables() |
| if 'raster_unit' not in var.op.name and 'VGG16' not in var.op.name] |
| gvs = self.optimizer.compute_gradients(total_cost, var_list=tvars) |
| return total_cost, gvs |
|
|
| def build_training_op(self, grad_list): |
| with tf.variable_scope('train_op', reuse=tf.AUTO_REUSE): |
| gvs = self.average_gradients(grad_list) |
| g = self.hps.grad_clip |
|
|
| for grad, var in gvs: |
| print('>>', var.op.name) |
| if grad is None: |
| print(' >> None value') |
|
|
| capped_gvs = [(tf.clip_by_value(grad, -g, g), var) for grad, var in gvs] |
|
|
| self.train_op = self.optimizer.apply_gradients( |
| capped_gvs, global_step=self.global_step, name='train_step') |
|
|
| def average_gradients(self, grads_list): |
| """ |
| Compute the average gradients. |
| :param grads_list: list(of length N_GPU) of list(grad, var) |
| :return: |
| """ |
| avg_grads = [] |
| for grad_and_vars in zip(*grads_list): |
| grads = [] |
| for g, _ in grad_and_vars: |
| expanded_g = tf.expand_dims(g, 0) |
| grads.append(expanded_g) |
| grad = tf.concat(grads, axis=0) |
| grad = tf.reduce_mean(grad, axis=0) |
|
|
| v = grad_and_vars[0][1] |
| grad_and_var = (grad, v) |
| avg_grads.append(grad_and_var) |
|
|
| return avg_grads |