{"componentChunkName":"component---src-templates-acg-portal-new-template-tsx","path":"/lmpb58sbb","result":{"data":{"markdownRemark":{"html":"

简介

\n

VideoQualityScore 是基于 Daft + Ray 分布式框架的视频质量评分组件，利用 CLIP 模型对视频帧进行质量评估，输出一个综合质量分数，适用于视频筛选、质量过滤等场景。视频质量评分处理器，基于 CLIP-IQA 对采样帧评分并聚合为视频质量分

\n

功能

\n\n

输入

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
输入列名说明
video_paths可选，视频文件路径列（本地、BOS、HTTP等），与 video_binaries 二选一
video_binaries可选，视频二进制数据列，与 video_paths 二选一
video_formats可选，视频格式字符串列，配合 video_binaries 使用
\n

参数

\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n\n
参数名称类型默认值描述
model_pathstr/opt/aihc/models模型文件存储路径
clip_model_namestropenai/clip-vit-large-patch14CLIP 模型名称或路径
promptstrquality质量评估提示词
devicestrcuda设备类型，支持CPU和GPU设备默认值：\"cuda\"可选值：\"cpu\", \"cuda\", \"cuda:0\", \"cuda:1\"等
sample_modestrby_count_uniform采样模式，参考 VideoFrameSampler
start_time_secfloat0.0采样起始时间（秒）
end_time_secfloat or None采样结束时间（秒）
count_kint or None8均匀采样帧数（by_count_uniform使用）
interval_secfloat or None时间间隔（秒，by_interval_time使用）
interval_framesint or None解码帧间隔（by_interval_frames使用）
target_fpsfloat or None目标采样FPS（by_fps使用）
timestamps_seclist[float] or None采样时间戳列表（秒，by_timestamps使用）
max_framesint or None返回帧上限
reduce_modestravg多帧聚合策略，可选 \"avg\"
default_video_formatstrmp4二进制/BASE64输入的视频格式
\n

输出

\n

包含视频质量评分的浮点数组，无法评分返回null值

\n

使用示例

\n\n
\n
\n
\n Plain Text\n \n
\n
\n 
1from __future__ import annotations\n2\n3import logging\n4import os\n5\n6import ray\n7\n8import daft\n9from daft import col\n10from daft.aihc.common.udf import aihc_udf\n11from daft.aihc.functions.video.video_quality_score import VideoQualityScore\n12\n13if __name__ == "__main__":\n14    if os.getenv("DAFT_RUNNER", "native") == "ray":\n15        import ray\n16        ray.init(dashboard_host="0.0.0.0", ignore_reinit_error=True)\n17        daft.set_runner_ray()\n18    daft.set_execution_config(actor_udf_ready_timeout=6000, min_cpu_per_task=0)\n19\n20    model_path = os.getenv("MODEL_PATH", "/opt/aihc/models")\n21    num_gpus = 1\n22\n23    samples = {\n24        "video_path": ["https://{bucket}.bj.bcebos.com/sample.mp4"],\n25    }\n26    ds = daft.from_pydict(samples)\n27\n28    constructor_kwargs = {\n29        "model_path": model_path,\n30        "clip_model_name": "openai/clip-vit-large-patch14",\n31        "prompt": "quality",\n32        "rank": 0,\n33        "sample_mode": "by_count_uniform",\n34        "count_k": 4,\n35        "reduce_mode": "avg",\n36    }\n37\n38    ds = ds.with_column(\n39        "quality_score",\n40        aihc_udf(VideoQualityScore, construct_args=constructor_kwargs, num_gpus=num_gpus, batch_size=1, concurrency=1)(\n41            col("video_path")\n42        ),\n43    )\n44\n45    ds.show()\n46#╭────────────────────────────────┬───────────────────╮                                                                                                                                                         \n47#│ video_path                     ┆ quality_score     │\n48#│ ---                            ┆ ---               │\n49#│ String                         ┆ Float64           │\n50#╞════════════════════════════════╪═══════════════════╡\n51#│ https://{bucket}.bj.bcebos.com/┆ 0.372357835372289 │\n52#╰────────────────────────────────┴───────────────────╯
\n
\n
\n
\n ","fields":{"slug":"lmpb58sbb","title":"视频质量评分","date":"2026-06-04","extractedHeadings":[]},"headings":[{"value":"简介","depth":2},{"value":"功能","depth":2},{"value":"输入","depth":2},{"value":"参数","depth":2},{"value":"输出","depth":2},{"value":"使用示例","depth":2}]}},"pageContext":{"isCreatedByStatefulCreatePages":false,"slug":"lmpb58sbb","prev":{"id":"smpb58s91","name":"视频片段切分-时长","path":"smpb58s91","filePath":"操作指南/AI数据处理/算子列表/视频/视频片段切分-时长.md","seo":null,"parentIds":["ilib2qygp","Ymo88m8hi","Imob3m6so","6mpawyry0"],"parents":[{"id":"ilib2qygp","documentId":"bfa43a8b-968a-41a1-8c9d-906507eeaed9","name":"操作指南","repoName":"AIHC","filePath":"操作指南","disabled":false,"path":"ilib2qygp","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":null},{"id":"Ymo88m8hi","documentId":"c8cb5e38-f8c5-40f4-a424-b0c7895f0c0a","name":"AI数据处理","repoName":"AIHC","filePath":"操作指南/AI数据处理","disabled":false,"path":"Ymo88m8hi","lastMergeTime":"2026-04-21 14:23:10","isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"Imob3m6so","documentId":"fe548e34-6659-4ff5-86f6-eee2c43aec90","name":"算子列表","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表","disabled":false,"path":"Imob3m6so","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"6mpawyry0","documentId":"3428d02c-e287-45b8-9d16-5a7b20881dd4","name":"视频","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表/视频","disabled":false,"path":"6mpawyry0","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""}]},"next":{"id":"vmpb58sdc","name":"通用视频格式转换","path":"vmpb58sdc","filePath":"操作指南/AI数据处理/算子列表/视频/通用视频格式转换.md","seo":null,"parentIds":["ilib2qygp","Ymo88m8hi","Imob3m6so","6mpawyry0"],"parents":[{"id":"ilib2qygp","documentId":"bfa43a8b-968a-41a1-8c9d-906507eeaed9","name":"操作指南","repoName":"AIHC","filePath":"操作指南","disabled":false,"path":"ilib2qygp","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":null},{"id":"Ymo88m8hi","documentId":"c8cb5e38-f8c5-40f4-a424-b0c7895f0c0a","name":"AI数据处理","repoName":"AIHC","filePath":"操作指南/AI数据处理","disabled":false,"path":"Ymo88m8hi","lastMergeTime":"2026-04-21 14:23:10","isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"Imob3m6so","documentId":"fe548e34-6659-4ff5-86f6-eee2c43aec90","name":"算子列表","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表","disabled":false,"path":"Imob3m6so","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"6mpawyry0","documentId":"3428d02c-e287-45b8-9d16-5a7b20881dd4","name":"视频","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表/视频","disabled":false,"path":"6mpawyry0","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""}]},"parents":[{"id":"ilib2qygp","documentId":"bfa43a8b-968a-41a1-8c9d-906507eeaed9","name":"操作指南","repoName":"AIHC","filePath":"操作指南","disabled":false,"path":"ilib2qygp","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":null},{"id":"Ymo88m8hi","documentId":"c8cb5e38-f8c5-40f4-a424-b0c7895f0c0a","name":"AI数据处理","repoName":"AIHC","filePath":"操作指南/AI数据处理","disabled":false,"path":"Ymo88m8hi","lastMergeTime":"2026-04-21 14:23:10","isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"Imob3m6so","documentId":"fe548e34-6659-4ff5-86f6-eee2c43aec90","name":"算子列表","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表","disabled":false,"path":"Imob3m6so","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""},{"id":"6mpawyry0","documentId":"3428d02c-e287-45b8-9d16-5a7b20881dd4","name":"视频","repoName":"AIHC","filePath":"操作指南/AI数据处理/算子列表/视频","disabled":false,"path":"6mpawyry0","lastMergeTime":null,"isApiDoc":null,"httpMethod":null,"seo":null,"sourceOrgName":null,"sourceRepoName":null,"sourceDocumentId":""}],"specificSeo":null}}}