Spaces:
Runtime error
Runtime error
File size: 2,388 Bytes
e629302 |
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 |
import numpy as np
import cv2
import numexpr
from PIL import Image
def parse_keyframe_string(string, max_frames):
res = np.ones(max_frames)
parts = string.split(",")
keyframes = {}
for part in parts:
try:
k, v = part.split(":")
keyframes[int(k.strip())] = v.strip("() ")
except: continue
sorted_keys = sorted(keyframes.keys())
for i in range(len(sorted_keys)):
start_f = sorted_keys[i]
end_f = sorted_keys[i+1] if i+1 < len(sorted_keys) else max_frames
val_str = keyframes[start_f]
for f in range(start_f, end_f):
if val_str.replace('.','',1).isdigit():
res[f] = float(val_str)
else:
try:
res[f] = numexpr.evaluate(val_str, local_dict={'t': f, 'sin': np.sin, 'cos': np.cos, 'pi': np.pi}).item()
except:
res[f] = res[f-1] if f > 0 else 0.0
return res
def maintain_colors(prev_img, target_img):
"""Matches the color histogram of the new frame to the first frame/previous frame."""
prev_img_cv = cv2.cvtColor(np.array(prev_img), cv2.COLOR_RGB2LAB)
target_img_cv = cv2.cvtColor(np.array(target_img), cv2.COLOR_RGB2LAB)
avg_l, avg_a, avg_b = np.mean(prev_img_cv[:,:,0]), np.mean(prev_img_cv[:,:,1]), np.mean(prev_img_cv[:,:,2])
target_img_cv[:,:,0] = np.clip(target_img_cv[:,:,0] + (avg_l - np.mean(target_img_cv[:,:,0])), 0, 255)
target_img_cv[:,:,1] = np.clip(target_img_cv[:,:,1] + (avg_a - np.mean(target_img_cv[:,:,1])), 0, 255)
target_img_cv[:,:,2] = np.clip(target_img_cv[:,:,2] + (avg_b - np.mean(target_img_cv[:,:,2])), 0, 255)
return Image.fromarray(cv2.cvtColor(target_img_cv, cv2.COLOR_LAB2RGB))
def anim_frame_warp(img, angle, zoom, translation_x, translation_y):
width, height = img.size
center = (width // 2, height // 2)
matrix = cv2.getRotationMatrix2D(center, angle, zoom)
matrix[0, 2] += translation_x
matrix[1, 2] += translation_y
return Image.fromarray(cv2.warpAffine(np.array(img), matrix, (width, height), borderMode=cv2.BORDER_REPLICATE))
def lerp_frames(frame1, frame2, alpha):
arr1 = np.array(frame1).astype(np.float32)
arr2 = np.array(frame2).astype(np.float32)
blended = arr1 * (1 - alpha) + arr2 * alpha
return Image.fromarray(blended.astype(np.uint8)) |