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。

Five Learnings from 15 Years in Perception

67 点作者 reteltech7 个月前

4 条评论

snovv_crash6 个月前
It&#x27;s a fine balance between Not Invented Here syndrome vs. trying to hammer the square peg of off-the-shelf OSS into the round hole of the actual problem you&#x27;re trying to solve.<p>For example they suggest ROS as a robust industry-ready software, which absolutely hasn&#x27;t been my experience: you hire a bunch of domain experts to solve the various [hardware, controls, perception, imaging, systems] problems, but once you use ROS as your middleware you end up needing a bunch of ROS experts instead. This is due to the horrible build system, odd choice of defaults, instability under constrained resources, and how it inserts itself into everything. You end up needing more fine-grained control than ROS gives you to make an actually robust system, but by the time you discover this you&#x27;ll be so invested into ROS that switching away will involve a full rewrite.<p>The same goes for further downstream: OpenCV images are basically a void* with a bunch of helper functions. (4.x tried to help with this but got sideswiped by DNN before anything concrete could happen.)<p>I guess it&#x27;s the same rant the FreeBSD people have about the Linux ecosystem and its reliability. However I&#x27;d hope we raise our standards when it comes to mobile robotics that have the potential to accidentally seriously hurt people. And who knows, maybe one day OpenCV and ROS will pleasantly surprise me the way Linux has with its progress.
评论 #42136097 未加载
评论 #42134091 未加载
dghf6 个月前
Linguistic tangent: when did &quot;learnings&quot; oust &quot;lessons&quot; as the standard word for &quot;things I have learned&quot;?
评论 #42135754 未加载
评论 #42137633 未加载
DaiPlusPlus6 个月前
This article struck a personal note with me because around the same time (2008-2012) I was really getting into vision, and even got published as an undergrad for imaging sensor fusion work (...my first, only, and likely last only meaningful contribution to my species); while the wider MV&#x2F;CV community was making incremental gains every few years (anyone else remember Histogram-of-Oriented-Gradients?), that&#x27;s what they were: incremental (I also remember my research-supervisor recounting how the patent on SIFT probably held back the entire field by a decade or two, so yes - things were slow-moving...<p>...until a few years ago when:<p>&gt; Computer vision has been consumed by AI.<p>...but &quot;AI&quot; is an unsatisfying reduction. What does it even mean? (and c&#x27;mon, plenty of non-NN CV techniques going back decades can be called &quot;AI&quot; today with a straight-face (for example, an adaptive pixel+contour histogram model for classifying very specific things).<p>My point is that computer-vision, as a field, *is* (an) artificial-intelligence: it has not been &quot;consumed by AI&quot;. I don&#x27;t want ephemeral fad terminology (y&#x27;know... buzzwords) getting in the way of what could have been a much better article.
评论 #42133780 未加载
评论 #42136548 未加载
评论 #42134195 未加载
评论 #42142180 未加载
lynx236 个月前
Why is it that people working on spy tech never have an ethics section in there &quot;what I&#x27;ve learnt&quot; rumblings?
评论 #42133738 未加载