TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

DeepMind Shows AI Has Trouble Seeing Homer Simpson's Actions

3 点作者 the-mitr将近 8 年前

1 comment

pavement将近 8 年前
Oh come on, this is an easy one. They need to include the complete audio channel in their training model, and identify important audio cues.<p>99% of Homer&#x27;s characterization is really Dan Castellaneta&#x27;s voice. The rest is simply the basic stretch-and-squash rubber band concept, as applied to the bug-eyed, yellow Simpsons visual character model.<p><a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Rubber_hose_animation" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Rubber_hose_animation</a><p>This is not an image processing task.