Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Search results

  1. shrineofchance

    Tesla has allegedly activated a selfie cam Driver Monitoring System

    https://driveteslacanada.ca/model-y/tesla-officially-adds-driver-monitoring-system-through-cabin-camera-update/ archive: web.archive.org | archive.is
  2. shrineofchance

    Poll: your gut expectations for FSD v9

    Radical = Obvious, large, qualitative improvement Incremental = Only a bit better; about what you’d expect from a typical update Somewhere between = More than a typical update, but not enough improvement to make an obvious, striking contrast between v8.2 and v9
  3. shrineofchance

    Facebook AI: "Self-supervised learning: The dark matter of intelligence"

    Loved this blog post by Yann LeCun and Ishan Misra at Facebook AI. If self-supervised learning could be made to work as well for computer vision as for natural language processing (NLP), it would be a giant leap forward for robotics, including autonomous vehicles. Notably, Tesla is using...
  4. shrineofchance

    Google AI: Revisiting the Unreasonable Effectiveness of Data

    Why does the size of Tesla’s training fleet matter? It’s well over 1 million cars, with the fleets of all competitors worldwide amounting to a combined size of well under 10,000. But is data really that important? Can you really get much more with 1 million+ cars than with 100 or 1,000? Yes. A...
  5. shrineofchance

    4D vision

    This is 2D vision: Image credit: greentheonly. Projecting out 3D models from detections within individual 2D images. This is 3D vision: Image source: Karpathy at CVPR 2020. The neural network directly outputs a 3D model from the 8 camera images. This is 4D vision (3D + time)...
  6. shrineofchance

    The FSD v9 showstopper

    Projections from 2D images vs. native 3D top-down bird's eye view: Source: Andrej Karpathy at ScaledML in February 2020.
  7. shrineofchance

    Poll: when will FSD Beta v9.x go fleet-wide?

    You have 7 days to vote. You can change your vote until then. Votes will be displayed publicly.
  8. shrineofchance

    Autonomous Vehicles Powered By End-To-End Deep Learning with Alex Kendall, Wayve.ai

    https://pca.st/episode/0c96e03a-e68e-415c-8951-b8828c90585b
  9. shrineofchance

    Interview with Yann LeCun on AI (only quasi-related to autonomy)

    https://pca.st/episode/f3834ab5-5ae3-44de-8c2e-229bd1e12bea
  10. shrineofchance

    Interview with Andrej Karpathy about AI, autonomy, and Tesla

    https://pca.st/episode/8e024d93-90ab-4b72-b2f0-6e8971448bf4
  11. shrineofchance

    FSD Beta v9.0: “significant architectural changes“

    Elon and Karpathy have mentioned various fundamental changes to FSD over the past 18 months: HydraNet NN architecture multi-task learning 2D labelling to 3D labelling 2D labelling to 4D labelling: using video to incorporate the time dimension, rather than still images from per-camera imagery...
  12. shrineofchance

    Poll: what company is leading in driving automation?

    Poll will be open for 31 days (you can change your vote until then) and votes will be displayed publicly. Sharing your reasoning here is welcomed.
  13. shrineofchance

    Driver input is a label

    Tesla is going to see an explosion of training signal as the FSD Beta gets pushed fleet-wide. First, this will enhance data curation: when a disengagement occurs, there is a high probability that the sensors or vector space representations show a novel training example that can be added to a...
  14. shrineofchance

    NeurIPS paper: Causal Confusion in Imitation Learning

    Abstract: “Behavioral cloning reduces policy learning to supervised learning by training a discriminative model to predict expert actions given observations. Such discriminative models are non-causal: the training procedure is unaware of the causal structure of the interaction between the...
  15. shrineofchance

    Drago Anguelov – Machine Learning for Autonomous Driving at Scale

    Waymo's Head of Research on: simultaneous detection, tracking, and prediction representing the built environment and road users as polylines in vector space deep structure from motion (deep SfM) visual LIDAR (VIDAR) using generative adversarial networks (GANs) to generate synthetic data...
  16. shrineofchance

    How optimistic are you feeling about FSD progress in 2021?

    Not in terms of whether Tesla will achieve true Level 5 or anything like that, but whether you feel a lot of progress will happen between now and the end of the year.
  17. shrineofchance

    Jim Keller interview (February 2021)

    Check the video description for time codes.
  18. shrineofchance

    Posts stuck in moderation queue?

    I think some of my posts are getting stuck in the moderation queue because I'm a new user and the posts have Imgur links embedded in them. Requesting approval!
  19. shrineofchance

    Introduction to me

    Hello, I'm a Tesla fan and investor, located in North America. I currently don't own a Tesla because I'm scared of/bad at driving, so I hope they become fully self-driving soon.
  20. shrineofchance

    Driver monitoring & the selfie cam

    Some people claim the Tesla selfie cam can't do driver monitoring because it's too dark at night to see the driver's face. But the selfie cam is the same Aptina camera on the outside of the car. Here's a photo taken by greentheonly on an unlit street, using one of the side cameras: The tree...