EPISODE · Nov 10, 2024 · 3 MIN
HourVideo: 评估一小时视频语言理解能力的新基准数据集
from AI Podcast · host weedge
HourVideo 是一个新颖的基准数据集,旨在严格评估多模态模型对一小时视频的理解能力。该数据集包含一个新的任务套件,包括摘要、感知(回忆、跟踪)、视觉推理(空间、时间、预测、因果、反事实)和导航(房间到房间、对象检索)任务。HourVideo 包含来自 Ego4D 数据集的 500 个手动策划的以自我为中心的视频,时长从 20 分钟到 120 分钟不等,并包含 12,976 个高质量的五选一多项选择问题。
Embed this episode
Ready to play
HourVideo: 评估一小时视频语言理解能力的新基准数据集
0:00
3:13
1×
No transcript for this episode yet
Similar Episodes
No similar episodes found.
Similar Podcasts
No similar podcasts found.
Frequently Asked Questions
How long is this episode of AI Podcast?
This episode is 3 minutes long.
When was this AI Podcast episode published?
This episode was published on November 10, 2024.
Can I download this AI Podcast episode?
Yes. Use the download control on the episode player to save the publisher-provided media file.
URL copied to clipboard!