[CVPR'21 WAD] Keynote - Andrej Karpathy, Tesla
Channel:WAD at CVPR · Length:31:31 · Published:2021-06-24 · Views:51361
Summary
Andrej Karpathy, then Tesla's head of AI, gives a CVPR 2021 workshop keynote on how Autopilot's neural networks work. A large part covers the move to a vision-only stack, including why radar was dropped and how vision produces depth estimates.
Why HN recommends it
HN commenters cite it in Tesla threads as the primary-source explanation of why Tesla removed radar and how its vision depth estimation works.
This talk was worth watching (any Karpathy or Carmack talk is imo) — HN, 2023
Machine learning & deep learning · Self-driving & autonomous vehicles · Computer vision, speech & NLP
Subtitles (excerpt)
uh yes i'm very excited to be here thank you for the introduction um i would like to talk a bit more about what tesla has been up for the last few months um so first to start uh i guess we are here at the cvpr uh workshop on autonomous driving so i think i'm preaching to the core a little bit uh but i'd like to start with some slides on like why we're doing all of this and i think unfortunately we're kind of like in a bad shape when it comes to transportation in society and uh basically the issue is that we have these metallic objects traveling incredibly quickly and with very high kinetic energy and we are putting meat in the control loop in the into the system and so this is quite undesirable and really fundamentally what it comes down to is people are not very good at driving they get into a lot of trouble in accidents they don't want to drive and also in terms of economics we are involving people in transportation and of course we'd like to automate transportation and really reap the benefits of that automation in society so we all i think are on the same page that it should be possible to replace the meat computer with a silicon computer and get a lot of benefits out of it in terms of um safety and convenience and economics
On Hacker News
This talk was worth watching (any Karpathy or Carmack talk is imo): [VIDEO] Specifically this is the relevant part: [VIDEO] They removed radar because it was making things worse. — fossuser, 2023-05
Radar has poor resolution, which means it can't always differentiate between an object near the car's path and an object in the car's path. The radar can give false positives when there's a truck on the side of the road or a metal sign on an overpass. Then you have to use the vision cameras to figure out what's really going on. Radar also loses its "lock" sometimes, which can be an issue with car… — ggreer, 2022-05
> These are the same things that can cause a traffic accident for a human with human eyes, so there's really no logical reason to assume that you can create a more performing and safer computer driver by emulating human vision only. Okay. This is a concrete claim that I think is falsifiable. I'll give not just one, but multiple arguments. Argument 1: Connect Four, Chess, Checkers, Go, Limit Hold'… — JoshCole, 2022-05
If you're interested, Tesla's head of AI recently did a public presentation of their new vision based depth system. It's worth a watch if you find this stuff interesting and enjoy learning about the forefront of technology. [VIDEO] A key takeaway is that they've been running this stack in shadow mode (validating output but not controlling the car) on everyone's Tesla for quite some time. Equivale… — simondotau, 2021-07
He is talking about the long technical talk at CVPR. [VIDEO] For example look at this section: [VIDEO] Also if you look into it you will find Google's self-driving car project ran into similar issues. They upgraded the radar in response to the issues rather than removing it, but there are actual issues. Vision only was better than radar plus vision in their stack. For real world situations that f… — JoshCole, 2023-04
Related videos
- Keynote by John Carmack at Oculus Connect 2014 · kungfl · 1:36:59
- Susan Crawford on Captive Audience: The Telecom Industry & Monopoly Power in the New Gilded Age · Berkman Klein Center for Internet & Society · 1:05:24
- Hacking at Quantum Speed with Shor's Algorithm | Infinite Series · PBS Infinite Series · 16:36
- The effects of DBS on the motor symptoms of Parkinson's Disease · Andrew Johnson · 4:03
- Pranks Destroy Scam Callers- GlitterBomb Payback · Mark Rober · 26:40
- Tesla Autonomy Day · Tesla · 3:52:20
- Bret Victor - Inventing on Principle · Rui Oliveira · 54:20
- Alexander Pope Wrote a 35-Page Takedown of a Historian · CGP Grey · 21:32
- Andrej Karpathy - AI for Full-Self Driving at Tesla · Matroid · 30:08
- Epic v Apple: Judgment Day - Who Won? Who Lost? ...and Why? (VL538) · Hoeg Law · 2:26:21
- Starbase Tour with Elon Musk [PART 1 // Summer 2021] · Everyday Astronaut · 53:17
- ARCHIVE TEAM: A Distributed Preservation of Service Attack · textfilesdotcom · 44:18
- Dan Abramov: Beyond React 16 | JSConf Iceland · JSConf · 33:24
- The Stuff of Thought | Steven Pinker | Talks at Google · Talks at Google · 1:15:04
- Is the world getting better or worse? A look at the numbers | Steven Pinker · TED · 18:32
- The spelled-out intro to neural networks and backpropagation: building micrograd · Andrej Karpathy · 2:25:52