luni/cvpr
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Files
seq_pyr
cvpr/models/pyramid_model_optimized.py
T
luni eb3ad3b887 sequential pyramid
2023-03-11 22:53:30 +00:00

195 lines
9.5 KiB
Python
Raw Permalink Blame History

import os
import time
import random
import tensorflow as tf
import numpy as np
# import plotly.graph_objects as go
# import plotly.express as ex
from custom_utils.custom_utils import load_model
from models.base import AEEncoder, AEEncoderNew
from models.pyramid_level import PyramidLevel, PyramidLevelNew
from models.upscaler import UpscalerBasic, UpscalerBasicConv, UpscalerBasicNew
class PyramidNew(tf.keras.Model):
def __init__(self, config):
super(PyramidNew, self).__init__()
self.static_enc : AEEncoderNew = AEEncoderNew(config['cond_ae_static']['encoder_depths'], config['cond_ae_static']['encoder_dims'], config['cond_ae_static']['kernel_size'], config['cond_ae_static']['activation'])
if config['pyramid']['static_model'] != '':
load_model(self.static_enc, config['pyramid']['static_model'] + '_enc')
self.body_enc : AEEncoderNew = AEEncoderNew(config['cond_ae_body']['encoder_depths'], config['cond_ae_body']['encoder_dims'], config['cond_ae_body']['kernel_size'])
if config['pyramid']['body_model'] != '':
load_model(self.body_enc, config['pyramid']['body_model'] + '_enc')
if config['use_curvature']:
self.curv_enc : AEEncoderNew = AEEncoderNew(config['cond_ae_curv']['encoder_depths'], config['cond_ae_curv']['encoder_dims'], config['cond_ae_curv']['kernel_size'], config['cond_ae_curv']['activation'])
if config['pyramid']['curvature_model'] != '':
load_model(self.curv_enc, config['pyramid']['curvature_model'] + '_enc')
if config['pyramid']['freeze_encoders']:
self.static_enc.trainable = False
self.body_enc.trainable = False
if config['use_curvature']:
self.curv_enc.trainable = False
self.levels = [PyramidLevelNew({ **config['pyramid']['levels'][i] }) for i in range(len(config['pyramid']['levels']))]
self.add_noise = config['pyramid']['add_noise']
self.noise_amount = config['pyramid']['noise_amount']
self.use_offsets = config['use_offsets']
self.use_curvature = config['use_curvature']
self.upscale_enhancer = None
if config['pyramid']['final_upscale_enhancer'] != '':
k_size = config['pyramid']['final_upscale_enhancer_kernel_size'] if 'final_upscale_enhancer_kernel_size' in config['pyramid'].keys() else ((3 if config['pyramid']['resolutions'][-1] <= 64 else 5))
self.upscale_enhancer = UpscalerBasicConv(low_res=config['pyramid']['resolutions'][-1], kernel_size=k_size)
load_model(self.upscale_enhancer, config['pyramid']['final_upscale_enhancer'])
self.upscale_enhancer.trainable = False
def enhance(self, low, mask, training):
x = self.upscale_enhancer([low, mask], training=training)
return x
def encode_conds(self, uv_template, uv_body_posed, uv_curv, training):
z_static = self.static_enc(uv_template, training)
z_body = self.body_enc(uv_body_posed, training)
if self.use_curvature:
z_curv = self.curv_enc(uv_curv, training)
z_cond = tf.concat([z_static, z_body, z_curv], axis=-1)
else:
z_cond = tf.concat([z_static, z_body], axis=-1)
return z_cond
def sample(self, inputs):
levels_input, conds = inputs
uv_static, uv_body, uv_curv = conds[0], conds[1], conds[2]
z_cond = self.encode_conds(uv_static, uv_body, uv_curv, training=False)
pred = [None, None, None, None, None]
for i in range(len(self.levels)):
pred[i] = self.levels[i].sample(z_cond, levels_input[i][-2])
pred_1 = pred[0]
pred_2 = pred[1]
pred_3 = pred[2]
pred_4 = pred[3]
pred_5 = pred[4]
pred_1 = self.levels[0].enhance(pred_1, levels_input[0][-2], training=False)
pred_1 = tf.image.resize(pred_1, (self.levels[1].size, self.levels[1].size//2), tf.image.ResizeMethod.BILINEAR)
pred_1 = pred_1 * levels_input[1][-2][:,:,:,tf.newaxis]
pred_2 = pred_1 + pred_2
pred_2 = self.levels[1].enhance(pred_2, levels_input[1][-2], training=False)
pred_2 = tf.image.resize(pred_2, (self.levels[2].size, self.levels[2].size//2), tf.image.ResizeMethod.BILINEAR)
pred_2 = pred_2 * levels_input[2][-2][:,:,:,tf.newaxis]
pred_3 = pred_2 + pred_3
pred_3 = self.levels[2].enhance(pred_3, levels_input[2][-2], training=False)
pred_3 = tf.image.resize(pred_3, (self.levels[3].size, self.levels[3].size//2), tf.image.ResizeMethod.BILINEAR)
pred_3 = pred_3 * levels_input[3][-2][:,:,:,tf.newaxis]
pred_4 = pred_3 + pred_4
pred_4 = self.levels[3].enhance(pred_4, levels_input[3][-2], training=False)
pred_4 = tf.image.resize(pred_4, (self.levels[4].size, self.levels[4].size//2), tf.image.ResizeMethod.BILINEAR)
pred_4 = pred_4 * levels_input[4][-2][:,:,:,tf.newaxis]
pred_5 = pred_4 + pred_5
pred_5 = pred_5 * levels_input[4][-2][:,:,:,tf.newaxis]
return pred_5
def sample_seq(self, inputs):
levels_input, conds = inputs
uv_static, uv_body, uv_curv = conds[0], conds[1], conds[2]
z_cond = self.encode_conds(uv_static, uv_body, uv_curv, training=False)
# level 1
pred_1 = self.levels[0].sample(z_cond, levels_input[0][-2])
pred_1 = self.levels[0].enhance(pred_1, levels_input[0][-2], training=False)
pred_1 = tf.image.resize(pred_1, (self.levels[1].size, self.levels[1].size//2), tf.image.ResizeMethod.BILINEAR)
pred_1 = pred_1 * levels_input[1][-2][:,:,:,tf.newaxis]
# level 2
pred_2 = self.levels[1].sample(z_cond, levels_input[1][-2])
pred_2 = pred_1 + pred_2
pred_2 = self.levels[1].enhance(pred_2, levels_input[1][-2], training=False)
pred_2 = tf.image.resize(pred_2, (self.levels[2].size, self.levels[2].size//2), tf.image.ResizeMethod.BILINEAR)
pred_2 = pred_2 * levels_input[2][-2][:,:,:,tf.newaxis]
# level 3
pred_3 = self.levels[2].sample(z_cond, levels_input[2][-2])
pred_3 = pred_2 + pred_3
pred_3 = self.levels[2].enhance(pred_3, levels_input[2][-2], training=False)
pred_3 = tf.image.resize(pred_3, (self.levels[3].size, self.levels[3].size//2), tf.image.ResizeMethod.BILINEAR)
pred_3 = pred_3 * levels_input[3][-2][:,:,:,tf.newaxis]
# level 4
pred_4 = self.levels[3].sample(z_cond, levels_input[3][-2])
pred_4 = pred_3 + pred_4
pred_4 = self.levels[3].enhance(pred_4, levels_input[3][-2], training=False)
pred_4 = tf.image.resize(pred_4, (self.levels[4].size, self.levels[4].size//2), tf.image.ResizeMethod.BILINEAR)
pred_4 = pred_4 * levels_input[4][-2][:,:,:,tf.newaxis]
# level 5
pred_5 = self.levels[4].sample(z_cond, levels_input[4][-2])
pred_5 = pred_5 * levels_input[4][-2][:,:,:,tf.newaxis]
pred_5 = pred_4 + pred_5
return pred_5
def call(self, inputs, training):
levels_input, conds = inputs
uv_static, uv_body, uv_curv = conds[0], conds[1], conds[2]
z_cond = self.encode_conds(uv_static, uv_body, uv_curv, training)
if self.add_noise == True and training == True:
z_cond = z_cond + (tf.random.uniform(z_cond.shape, maxval=self.noise_amount * 2) - self.noise_amount)
out_levels = []
pred, mean, logvar, z_latent = self.levels[0](levels_input[0], z_cond, training=training)
if self.use_offsets:
pred = pred + levels_input[0][2] # uv_body_posed_32
if 1 != len(self.levels):
pred = self.levels[0].enhance(pred, levels_input[0][-2], training=training) # levels_input[i][-2]=uv_mask_res
pred = tf.image.resize(pred, (self.levels[1].size, self.levels[1].size//2), tf.image.ResizeMethod.BILINEAR)
out_levels.append([pred, mean, logvar])
pred = pred * levels_input[1][-2][:,:,:,tf.newaxis]
else:
# pred = self.enhance(pred, levels_input[0][-2], training=training)
pred = self.levels[0].enhance(pred, levels_input[0][-2], training=training) # levels_input[i][-2]=uv_mask_res
pred = tf.image.resize(pred, (512, 256), tf.image.ResizeMethod.BILINEAR)
pred = pred * levels_input[-1][-1][:,:,:,tf.newaxis] # full res uv mask
out_levels.append([pred, mean, logvar])
for i in range(1, len(self.levels)):
input = levels_input[i]
input[0] = input[0] - pred
pred_level, mean, logvar, z_latent = self.levels[i](input, z_cond, z_latent, training=training)
pred = pred + pred_level
if i != len(self.levels) - 1:
pred = self.levels[i].enhance(pred, levels_input[i][-2], training=training)
out_levels.append([pred, mean, logvar])
pred = tf.image.resize(pred, (self.levels[i+1].size, self.levels[i+1].size//2), tf.image.ResizeMethod.BILINEAR)
pred = pred * levels_input[i+1][-2][:,:,:,tf.newaxis]
else:
# if self.upscale_enhancer is not None:
# pred = self.enhance(pred, levels_input[-1][-2], training=training)
if len(self.levels) != 5:
pred = self.levels[-1].enhance(pred, levels_input[-1][-2], training=training)
pred = tf.image.resize(pred, (512, 256), tf.image.ResizeMethod.BILINEAR)
pred = pred * levels_input[-1][-1][:,:,:,tf.newaxis] # full res uv mask
out_levels.append([pred, mean, logvar])
return out_levels
Reference in New Issue View Git Blame Copy Permalink