使用 VideoDecoder 解码视频¶
在本例中,我们将学习如何使用 VideoDecoder
类解码视频。
首先,做一些准备工作:我们将从网上下载一个视频,并定义一个绘图工具。你可以忽略这部分,直接跳到下面的创建解码器。
from typing import Optional
import torch
import requests
# Video source: https://www.pexels.com/video/dog-eating-854132/
# License: CC0. Author: Coverr.
url = "https://videos.pexels.com/video-files/854132/854132-sd_640_360_25fps.mp4"
response = requests.get(url, headers={"User-Agent": ""})
if response.status_code != 200:
raise RuntimeError(f"Failed to download video. {response.status_code = }.")
raw_video_bytes = response.content
def plot(frames: torch.Tensor, title : Optional[str] = None):
try:
from torchvision.utils import make_grid
from torchvision.transforms.v2.functional import to_pil_image
import matplotlib.pyplot as plt
except ImportError:
print("Cannot plot, please run `pip install torchvision matplotlib`")
return
plt.rcParams["savefig.bbox"] = 'tight'
fig, ax = plt.subplots()
ax.imshow(to_pil_image(make_grid(frames)))
ax.set(xticklabels=[], yticklabels=[], xticks=[], yticks=[])
if title is not None:
ax.set_title(title)
plt.tight_layout()
创建解码器¶
现在我们可以从原始(编码)视频字节创建解码器。当然,你也可以使用本地视频文件并将路径作为输入,而不是下载视频。
from torchcodec.decoders import VideoDecoder
# You can also pass a path to a local file!
decoder = VideoDecoder(raw_video_bytes)
尚未被解码器解码,但我们已经可以通过 metadata
属性访问一些元数据,该属性是一个 VideoStreamMetadata
对象。
print(decoder.metadata)
VideoStreamMetadata:
duration_seconds_from_header: 13.8
begin_stream_seconds_from_header: 0.0
bit_rate: 505790.0
codec: h264
stream_index: 0
begin_stream_seconds_from_content: 0.0
end_stream_seconds_from_content: 13.8
width: 640
height: 360
num_frames_from_header: 345
num_frames_from_content: 345
average_fps_from_header: 25.0
duration_seconds: 13.8
begin_stream_seconds: 0.0
end_stream_seconds: 13.8
num_frames: 345
average_fps: 25.0
通过索引解码器来解码帧¶
first_frame = decoder[0] # using a single int index
every_twenty_frame = decoder[0 : -1 : 20] # using slices
print(f"{first_frame.shape = }")
print(f"{first_frame.dtype = }")
print(f"{every_twenty_frame.shape = }")
print(f"{every_twenty_frame.dtype = }")
first_frame.shape = torch.Size([3, 360, 640])
first_frame.dtype = torch.uint8
every_twenty_frame.shape = torch.Size([18, 3, 360, 640])
every_twenty_frame.dtype = torch.uint8
通过索引解码器将帧作为 torch.Tensor
对象返回。默认情况下,帧的形状是 (N, C, H, W)
,其中 N 是批大小,C 是通道数,H 是高度,W 是帧的宽度。批维度 N 仅在解码多于一帧时存在。可以使用 VideoDecoder
的 dimension_order
参数将维度顺序更改为 N, H, W, C
。帧的数据类型始终为 torch.uint8
。
注意
如果需要解码多个帧,我们建议改用批处理方法,因为它们速度更快:get_frames_at()
、get_frames_in_range()
、get_frames_played_at()
和 get_frames_played_in_range()
。下面将对此进行描述。
plot(first_frame, "First frame")

plot(every_twenty_frame, "Every 20 frame")

迭代帧¶
解码器是一个普通的 iterable(可迭代)对象,可以像这样进行迭代
for frame in decoder:
assert (
isinstance(frame, torch.Tensor)
and frame.shape == (3, decoder.metadata.height, decoder.metadata.width)
)
检索帧的 pts 和 duration¶
通过索引解码器返回纯粹的 torch.Tensor
对象。有时,获取有关帧的额外信息会很有用,例如它们的 pts (Presentation Time Stamp,演示时间戳) 和 duration(持续时间)。这可以通过 get_frame_at()
和 get_frames_at()
方法实现,它们将分别返回 Frame
和 FrameBatch
对象。
last_frame = decoder.get_frame_at(len(decoder) - 1)
print(f"{type(last_frame) = }")
print(last_frame)
type(last_frame) = <class 'torchcodec._frame.Frame'>
Frame:
data (shape): torch.Size([3, 360, 640])
pts_seconds: 13.76
duration_seconds: 0.04
other_frames = decoder.get_frames_at([10, 0, 50])
print(f"{type(other_frames) = }")
print(other_frames)
type(other_frames) = <class 'torchcodec._frame.FrameBatch'>
FrameBatch:
data (shape): torch.Size([3, 3, 360, 640])
pts_seconds: tensor([0.4000, 0.0000, 2.0000], dtype=torch.float64)
duration_seconds: tensor([0.0400, 0.0400, 0.0400], dtype=torch.float64)
plot(last_frame.data, "Last frame")
plot(other_frames.data, "Other frames")
Frame
和 FrameBatch
都包含 data
字段,其中包含解码后的张量数据。它们还包含 pts_seconds
和 duration_seconds
字段,对于 Frame
来说是单个整数,对于 FrameBatch
来说是 1-D torch.Tensor
(批处理中的每帧一个值)。
使用基于时间的索引¶
到目前为止,我们根据索引检索了帧。我们还可以使用 get_frame_played_at()
和 get_frames_played_at()
根据帧的播放时间来检索,这两个方法也分别返回 Frame
和 FrameBatch
。
frame_at_2_seconds = decoder.get_frame_played_at(seconds=2)
print(f"{type(frame_at_2_seconds) = }")
print(frame_at_2_seconds)
type(frame_at_2_seconds) = <class 'torchcodec._frame.Frame'>
Frame:
data (shape): torch.Size([3, 360, 640])
pts_seconds: 2.0
duration_seconds: 0.04
other_frames = decoder.get_frames_played_at(seconds=[10.1, 0.3, 5])
print(f"{type(other_frames) = }")
print(other_frames)
type(other_frames) = <class 'torchcodec._frame.FrameBatch'>
FrameBatch:
data (shape): torch.Size([3, 3, 360, 640])
pts_seconds: tensor([10.0800, 0.2800, 5.0000], dtype=torch.float64)
duration_seconds: tensor([0.0400, 0.0400, 0.0400], dtype=torch.float64)
plot(frame_at_2_seconds.data, "Frame played at 2 seconds")
plot(other_frames.data, "Other frames")
脚本总运行时间: (0 分 2.842 秒)