Question

目标：结尾的每个帧都应该有一个完整的sudowoodo。任何不是sudowoodo的东西都应该在gif中是透明的（我也可以用白色）。我的最终目标是要重新着色，所以我必须能够区分背景和宠物小精灵。

我正在使用从网站获取gif的请求，然后通过枕头获取框架。当我浏览框架时，伪影（主要是背景）正在干扰我的处置方法。我用https://ezgif.com/split检查它的外观。

我将拼贴的处理范围，框架，处理方法和盒子放在拼贴的周围，以帮助查看拼贴的堆叠方式（如下图所示）。我在底部放置了示例代码，这些代码将显示重构的帧。

那我如何用枕头正确地获得干净的镜框？

图片

注意：

对于枕头框架，索引号4在体内出现白线，框架索引号27具有绿色背景。并且多个帧在球体/分支中缺少绿色。
要使ezgif.com看到我发布的图片，您必须放置“忽略优化”，但我想要的最终照片是网站选项“重画每帧并包含上一帧的详细信息”中显示的内容。 / li>

使用的Gif：https://play.pokemonshowdown.com/sprites/xyani/sudowoodo.gif

枕头结果：

Ezgif的结果：

代码

Python版本
Python == 3.6.4
枕头== 7.0.0
请求== 2.23.0

from PIL import Image, ImageDraw, ImageFont
from io import BytesIO
import requests

#Depends on the dispose_method and disposal_extent will process accordingly
def method_dispose(i, frames, previous_frame):
    # 0 PIL = Overlay and pass
    # 1 PIL = Overlay and return previous
    # 2 PIL = Erase Overlay
    new_frame = previous_frame.copy()
    current_frame = frames.convert('RGBA')
    new_frame.alpha_composite(current_frame, dest=frames.dispose_extent[0:2], source=frames.dispose_extent)
    if frames.disposal_method is 0:
        return new_frame, Image.new('RGBA', box=frames.size)
    elif frames.disposal_method is 1:
        return new_frame, new_frame.copy()
    elif frames.disposal_method is 2:
        draw = ImageDraw.Draw(previous_frame)
        draw.rectangle(frames.dispose_extent, fill=(255, 255, 255, 0)) #fill white transparent
        return new_frame, previous_frame.copy()

# Goes through the frames and pastes them next to each other then shows
def simpleCollage(frames, num_images_width : int = 5, num_images_height : int = 10):
    width, height = frames.size
    compilation = Image.new('RGBA', size=(width * num_images_width, height * num_images_height))
    fnt = ImageFont.load_default().font
    for i in range(frames.n_frames):
        frames.seek(i)
        the_frame = frames.convert('RGBA')
        draw = ImageDraw.Draw(the_frame)
        draw.rectangle(frames.dispose_extent, outline=(255,173,0,255))
        draw.text((0,0), f"F{i}-M{frames.disposal_method}", font=fnt, fill=(255, 0, 0))
        compilation.paste(the_frame, box=(width * int(i % num_images_width), height * int(i / num_images_width)))
        if i == (num_images_width * num_images_height):
            break;
    compilation.show()

response = requests.get("https://play.pokemonshowdown.com/sprites/xyani/sudowoodo.gif")
frames = Image.open(BytesIO(response.content))
simpleCollage(frames)

width, height = frames.size
all_frames = []
pass_frame = Image.new('RGBA', size=frames.size)
for i in range(frames.n_frames):
    frames.seek(i)
    disp_frame, pass_frame = method_dispose(i, frames, pass_frame)
    all_frames.append(disp_frame)

all_frames[0].save(fp="test.gif", format='GIF', save_all=True, append_images=all_frames[1:], optimize=False, duration=frames.info['duration'], loop=0)
simpleCollage(Image.open("test.gif"))

我用于处置方法的一些来源：

Answer 1

所产生的不良图像是由重复的.convert()引起的。为了解决这个问题，需要将.seek(0)放在.convert()之后。因此，结果代码应如下所示：

for i in range(frames.n_frames):
    frames.seek(i)
    disp_frame, pass_frame = method_dispose(i, frames, pass_frame)
    all_frames.append(disp_frame)
    frames.seek(0) # <-- Added

Answer 2

这不是经过充分测试，可以完全解决的答案，但是我想保存并分享正在进行的工作，这可能对其他查看此问题的人有用。

您可以使用 ImageMagick 来检查GIF中帧的范围，处置，延迟等，

identify -format "%f[%s] canvas=%Wx%H size=%wx%h offset=%X%Y %D %Tcentisecs\n" treething.gif

示例输出

treething.gif[0] canvas=58x66 size=58x66 offset=+0+0 Background 3centisecs
treething.gif[1] canvas=58x66 size=58x66 offset=+0+0 None 3centisecs
treething.gif[2] canvas=58x66 size=49x63 offset=+7+1 None 3centisecs
treething.gif[3] canvas=58x66 size=55x47 offset=+2+1 Background 3centisecs
treething.gif[4] canvas=58x66 size=56x64 offset=+2+1 Background 3centisecs
treething.gif[5] canvas=58x66 size=56x64 offset=+2+1 Background 3centisecs
treething.gif[6] canvas=58x66 size=56x64 offset=+2+1 Background 3centisecs
treething.gif[7] canvas=58x66 size=56x64 offset=+2+1 Background 3centisecs
treething.gif[8] canvas=58x66 size=56x64 offset=+2+1 None 3centisecs
treething.gif[9] canvas=58x66 size=54x65 offset=+3+1 Background 3centisecs
treething.gif[10] canvas=58x66 size=54x65 offset=+3+1 Background 3centisecs
treething.gif[11] canvas=58x66 size=54x65 offset=+3+1 None 3centisecs
treething.gif[12] canvas=58x66 size=45x58 offset=+7+6 Background 3centisecs
treething.gif[13] canvas=58x66 size=51x58 offset=+7+6 Background 3centisecs
treething.gif[14] canvas=58x66 size=55x64 offset=+2+0 Background 3centisecs
treething.gif[15] canvas=58x66 size=55x64 offset=+2+0 Background 3centisecs
treething.gif[16] canvas=58x66 size=55x64 offset=+2+0 Background 3centisecs
treething.gif[17] canvas=58x66 size=55x64 offset=+2+0 None 3centisecs
treething.gif[18] canvas=58x66 size=41x53 offset=+16+4 Background 3centisecs
treething.gif[19] canvas=58x66 size=56x60 offset=+1+4 None 3centisecs
treething.gif[20] canvas=58x66 size=19x61 offset=+9+3 None 3centisecs
treething.gif[21] canvas=58x66 size=43x54 offset=+8+10 None 3centisecs
treething.gif[22] canvas=58x66 size=43x47 offset=+8+10 None 3centisecs
treething.gif[23] canvas=58x66 size=50x61 offset=+1+3 Background 3centisecs
treething.gif[24] canvas=58x66 size=51x61 offset=+1+3 None 3centisecs
treething.gif[25] canvas=58x66 size=43x50 offset=+7+4 None 3centisecs
treething.gif[26] canvas=58x66 size=47x65 offset=+10+1 Background 3centisecs
treething.gif[27] canvas=58x66 size=50x65 offset=+7+1 None 3centisecs
treething.gif[28] canvas=58x66 size=47x64 offset=+3+0 Background 3centisecs
treething.gif[29] canvas=58x66 size=56x64 offset=+2+0 Background 3centisecs
treething.gif[30] canvas=58x66 size=58x64 offset=+0+0 Background 3centisecs
treething.gif[31] canvas=58x66 size=58x64 offset=+0+0 Background 3centisecs
treething.gif[32] canvas=58x66 size=58x64 offset=+0+0 None 3centisecs
treething.gif[33] canvas=58x66 size=37x59 offset=+15+2 Background 3centisecs
treething.gif[34] canvas=58x66 size=48x63 offset=+7+1 Background 3centisecs
treething.gif[35] canvas=58x66 size=48x63 offset=+7+1 Background 3centisecs
treething.gif[36] canvas=58x66 size=50x63 offset=+7+1 Background 3centisecs
treething.gif[37] canvas=58x66 size=50x63 offset=+7+1 Background 3centisecs
treething.gif[38] canvas=58x66 size=50x63 offset=+7+1 None 3centisecs
treething.gif[39] canvas=58x66 size=52x65 offset=+1+1 Background 3centisecs
treething.gif[40] canvas=58x66 size=57x65 offset=+1+1 Background 3centisecs
treething.gif[41] canvas=58x66 size=57x65 offset=+1+1 None 3centisecs
treething.gif[42] canvas=58x66 size=53x63 offset=+1+2 Background 3centisecs
treething.gif[43] canvas=58x66 size=54x63 offset=+1+2 Background 3centisecs
treething.gif[44] canvas=58x66 size=54x65 offset=+1+0 Background 3centisecs
treething.gif[45] canvas=58x66 size=55x65 offset=+1+0 Background 3centisecs
treething.gif[46] canvas=58x66 size=56x65 offset=+1+0 None 3centisecs
treething.gif[47] canvas=58x66 size=44x54 offset=+14+10 Background 3centisecs
treething.gif[48] canvas=58x66 size=44x63 offset=+14+2 Background 3centisecs
treething.gif[49] canvas=58x66 size=44x63 offset=+14+2 Background 3centisecs
treething.gif[50] canvas=58x66 size=44x63 offset=+14+2 Background 3centisecs
treething.gif[51] canvas=58x66 size=51x63 offset=+7+2 Background 3centisecs
treething.gif[52] canvas=58x66 size=51x63 offset=+7+2 None 3centisecs
treething.gif[53] canvas=58x66 size=40x31 offset=+14+3 Background 3centisecs
treething.gif[54] canvas=58x66 size=52x55 offset=+4+2 Background 3centisecs
treething.gif[55] canvas=58x66 size=52x55 offset=+4+2 None 3centisecs
treething.gif[56] canvas=58x66 size=49x51 offset=+1+6 None 3centisecs
treething.gif[57] canvas=58x66 size=56x63 offset=+2+2 Background 3centisecs
treething.gif[58] canvas=58x66 size=56x65 offset=+2+0 None 3centisecs
treething.gif[59] canvas=58x66 size=48x64 offset=+7+1 None 3centisecs

您还可以在draw.text((0,0)...)之后将以下行插入到程序中，以获得类似的输出：

print(i,frames.dispose_extent, frames.disposal_method)

现在我要做的就是比较两个...

我仍在尝试找出问题所在-我的怀疑是：

您需要在compilation.paste(the_frame,...)行中添加一个掩码，或者
the_frame = frames.convert('RGBA')对您不利，因为GIF实际上已被淡化，并且转换出现问题。

您还可以在红色背景上查看来自 ImageMagick 的合并帧：

magick tree.gif miff:- | magick montage -background red -geometry +5+5 -tile 12x miff:- montage.png

Python枕头Gif物件/背景

图片

代码

2 个答案: