Top all-time · #5799
Tombone's Computer Vision Blog
Personal blog · computervisionblog.comaiprogramming
Tombone's Computer Vision Blog is a personal blog at computervisionblog.com about ai and programming. Since 2015, 14 different Hacker News readers have linked it 15 times, most often “The Future of Real-Time SLAM and “Deep Learning vs. SLAM””. The last post was in 2019, but the archive is still widely read.
Too few sampled comments (8) to say how often HN recommends it.?
Why HN links it
HN readers link this computer vision blog for technical posts on topics such as SLAM and deep learning versus probabilistic methods. Most links are cited as background or support for a point.
Summarised by Claude from a sample of 8 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
A Blog about Deep Learning, Computer Vision, and the algorithms that are shaping the future of Artificial Intelligence.
Most-linked posts?
- The Future of Real-Time SLAM and “Deep Learning vs. SLAM”3 people · 3 mentions · first linked 2016-01-19
- Deep Learning vs. Probabilistic Graphical Models vs. Logic3 people · 3 mentions · first linked 2015-05-28
- Deep Learning Trends ICLR 20162 people · 2 mentions · first linked 2016-06-01
What Hacker News says?
Here's a very useful blog that deals exactly with this topic, http://www.computervisionblog.com/2016/01/why-slam-matters-f... I'd advise others on this thread, who tend to dismiss the difficulty of vision, to realize that techniques in ML and Deep-learning are very problem specific and not the holy grail of AGI they're made out to be.
Comment linking /2016/01/why-slam-matters-future-of-real-time.html
on “All Tesla Cars Being Produced Now Have Full Self-Driving Hardware”Submitted The Future of Real-Time SLAM and “Deep Learning vs. SLAM” as a story · 60 points
HN discussionGraphical models[0] & probabilistic programming[1], with the latter making it easier for developers to dive into this growing AI trend. Research in the field for the past decade has been steadily booming with more companies like Microsoft leading the way. I recommend checking out some MOOCs[2] in coursera. [0] http://www.computervisionblog.com/2015/04/deep-learning-vs-p... [1] http://probabilistic-programming.org/wi…
Comment linking /2015/04/deep-learning-vs-probabilistic.html
on “Ask HN: What are the most interesting emerging fields in computer science?”Submitted Deep down the rabbit hole: CVPR 2015 and beyond as a story · 9 points
HN discussionI actually strongly agree with him, In fact the progress in Deep Learning has stalled due to inefficiencies in sharing data and code between researchers. Today to replicate a study you have to spend several days downloading data, making sure all dependencies are installed etc. This is worsened by lack of standard formats for even simple things such as representing 2D bounding box on an image. People forget that the…
Comment linking /2015/01/from-feature-descriptors-to-deep.html
on “Google Brain Residency”Submitted Nuts and Bolts of Building Deep Learning Applications: NIPS 2016 as a story · 4 points
HN discussionFrom a survey of Deep Learning published in Nature[1]: "The issue of representation lies at the heart of the debate between the logic-inspired and the neural-network-inspired paradigms for cognition. In the logic-inspired paradigm, an instance of a symbol is something for which the only property is that it is either identical or non-identical to other symbol instances. It has no internal structure that is relevant t…
Comment linking /2015/04/deep-learning-vs-probabilistic.html
on “King – man + woman is queen; but why?”Submitted Deep Learning Trends ICLR 2016 as a story · 2 points
HN discussionI think the question is a little too unspecific for there to be a good answer. The field is vast and depending on which thing in computer vision you want to tackle the best learning paths may vary greatly. Just to give a bit of an overview: Before the Deep Learning Craze started in 2011 more classical Machine Learning techniques were used in CV: Support Vector Machines, Boosting, Decision Trees, etc.. These were (an…
Comment linking /
on “Ask HN: What are the best resources to learn computer vision?”Submitted Deep Learning Trends at ICLR 2016 as a story · 2 points
HN discussionThe article is typical clickbait, with little content let alone nuanced analysis. For those looking for a quick technical overview of SLAM take a look at blog post (by now head of vision at Magic Leap) http://www.computervisionblog.com/2016/01/why-slam-matters-f... ARkit looks like a prelude to AR glasses. Where given the ability to control hardware specs, the advantage is not transferable.
Comment linking /2016/01/why-slam-matters-future-of-real-time.html?m=1
on “Apple’s AR is closer to reality than Google’s”Submitted Deep Learning vs. Probabilistic Graphical Models vs. Logic as a story · 1 points
HN discussionMy layman's understanding is that it's mostly deep learning these days (what a surprise, I'm sure). For example, this article says that deep learning is behind most of Mobileye's computer vision stuff: http://www.computervisionblog.com/2015/03/mobileyes-quest-to... (They're the ones who did Tesla's original Autopilot hardware, although not the new stuff they just started shipping.)
Comment linking /2015/03/mobileyes-quest-to-put-deep-learning.html
on “Simple Lane Detection”Submitted Deep Learning vs. Big Data: Who Owns What? as a story · 1 points
HN discussionRelated: > Today’s SLAM systems help machines geometrically understand the immediate world (i.e., build associations in a local coordinate system) while today’s Deep Learning systems help machines reason categorically (i.e., build associations across distinct object instances). In conclusion, I share Newcombe and Davison excitement in Visual SLAM, as vision-based algorithms are going to turn Augmented and Virtual Re…
Comment linking /2016/01/why-slam-matters-future-of-real-time.html
on “Iris Automation (YC S16) gives drones situational awareness to fly autonomously”
