Spaces:
Runtime error
Runtime error
File size: 3,932 Bytes
441b0c8 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 |
import numpy as np
import cv2
import numexpr
import re
from PIL import Image
# --- Math Parsing ---
def parse_weight_string(string, max_frames):
"""Parses schedule strings with math support (e.g., '0:(0.5), 50:(sin(t/10))')."""
string = re.sub(r'\s+', '', str(string))
keyframes = {}
parts = string.split(',')
for part in parts:
try:
if ':' not in part: continue
frame_str, val_str = part.split(':', 1)
keyframes[int(frame_str)] = val_str.strip('()')
except: continue
if 0 not in keyframes: keyframes[0] = "0"
sorted_frames = sorted(keyframes.keys())
series = np.zeros(int(max_frames))
for i in range(len(sorted_frames)):
f_start = sorted_frames[i]
f_end = sorted_frames[i+1] if i < len(sorted_frames)-1 else int(max_frames)
formula = keyframes[f_start]
for f in range(f_start, f_end):
t = f
try:
val = numexpr.evaluate(formula, local_dict={'t': t, 'pi': np.pi, 'sin': np.sin, 'cos': np.cos, 'tan': np.tan})
series[f] = float(val)
except:
try: series[f] = float(formula)
except: series[f] = series[f-1] if f > 0 else 0.0
return series
# --- Image Processing ---
def get_border_mode(mode_str):
return {
'Reflect': cv2.BORDER_REFLECT_101,
'Replicate': cv2.BORDER_REPLICATE,
'Wrap': cv2.BORDER_WRAP,
'Black': cv2.BORDER_CONSTANT
}.get(mode_str, cv2.BORDER_REFLECT_101)
def maintain_colors(prev_img, color_match_sample, mode='LAB'):
"""Matches colors using LAB or HSV space to prevent drift."""
if mode == 'None' or prev_img is None or color_match_sample is None: return prev_img
prev_np = np.array(prev_img).astype(np.uint8)
sample_np = np.array(color_match_sample).astype(np.uint8)
if mode == 'LAB':
prev_lab = cv2.cvtColor(prev_np, cv2.COLOR_RGB2LAB)
sample_lab = cv2.cvtColor(sample_np, cv2.COLOR_RGB2LAB)
for i in range(3): # Match L, A, and B channels
avg_p = np.mean(prev_lab[:,:,i])
avg_s = np.mean(sample_lab[:,:,i])
prev_lab[:,:,i] = np.clip(prev_lab[:,:,i] - avg_p + avg_s, 0, 255)
return Image.fromarray(cv2.cvtColor(prev_lab, cv2.COLOR_LAB2RGB))
elif mode == 'HSV':
prev_hsv = cv2.cvtColor(prev_np, cv2.COLOR_RGB2HSV)
sample_hsv = cv2.cvtColor(sample_np, cv2.COLOR_RGB2HSV)
# Match Saturation and Value only, keep Hue
for i in [1, 2]:
avg_p = np.mean(prev_hsv[:,:,i])
avg_s = np.mean(sample_hsv[:,:,i])
prev_hsv[:,:,i] = np.clip(prev_hsv[:,:,i] - avg_p + avg_s, 0, 255)
return Image.fromarray(cv2.cvtColor(prev_hsv, cv2.COLOR_HSV2RGB))
return prev_img
def add_noise(img, noise_amt):
"""Adds uniform noise for texture injection."""
if noise_amt <= 0 or img is None: return img
img_np = np.array(img).astype(np.float32)
noise = np.random.normal(0, noise_amt * 255, img_np.shape).astype(np.float32)
noisy_img = np.clip(img_np + noise, 0, 255).astype(np.uint8)
return Image.fromarray(noisy_img)
def anim_frame_warp_2d(prev_img_pil, args_dict, border_mode_str='Reflect'):
"""Performs 2D affine transformation."""
if prev_img_pil is None: return None
cv2_img = np.array(prev_img_pil)
height, width = cv2_img.shape[:2]
center = (width // 2, height // 2)
angle = args_dict.get('angle', 0)
zoom = args_dict.get('zoom', 1.0)
tx = args_dict.get('translation_x', 0)
ty = args_dict.get('translation_y', 0)
trans_mat = cv2.getRotationMatrix2D(center, angle, zoom)
trans_mat[0, 2] += tx
trans_mat[1, 2] += ty
border_mode = get_border_mode(border_mode_str)
warped = cv2.warpAffine(cv2_img, trans_mat, (width, height), borderMode=border_mode)
return Image.fromarray(warped) |