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。

Homography Explained with Code

45 点作者 diginova6 个月前

3 条评论

collingreen6 个月前
It&#x27;s interesting how there are so many papers published in this space but they all tend to use these same few images (the webcam chessboard, the two buildings, and the diagrams). I&#x27;ve been looking into how to reliably stitch video frames into &quot;orthographs&quot; (similar problem is faced by aerial surveys; there is a lot of good work on this from the drone community) recently and have read probably two dozen recent papers spanning homography, photogrammetry, feature detection, sfm, nerf, and segmentation and most of them reuse these diagrams and at least some of these images.<p>Maybe the world would benefit from some more well documented, open licensed training&#x2F;validation data?
评论 #42314220 未加载
评论 #42315252 未加载
amstan6 个月前
&gt; Multiple View Geometry in Computer Vision, Richard Hartley and Andrew Zisserman, [117] (some sample chapters are available here, CVPR Tutorials are available here)<p>Heh, about 10 years ago I read that book and figured out a few things:<p>* triangulate position in 3d space of an object given a few 2d pictures (from known camera locations, ie: camera intrinsic and extrinsic matrices)<p>* how to instruct my friend to make a simple 3d rendering &quot;engine&quot;<p>Nice to see the same stuff distilled into an article.
Neeloppher6 个月前
Intresting