←
WAC2022665 posts
WAC2022 avatar

WAC2022

@WebAudioConf

#WAC2022 wac2022.i3s.univ-cotedazur.f…

◷ Joined September 2014▧ 197 Photos and videos
wiggle avatar
⟳ WAC2022 retweeted
wiggle@w1991e·7 May 2024

Robot factory in China 🇨🇳

0:24
SkalskiP avatar
⟳ WAC2022 retweeted
SkalskiP@skalskip92·21 Feb 2024

The YOLO-World YouTube tutorial is out! please, let us know what you think! - model architecture - processing images and video in Colab - prompt engineering and detection refinement - pros and cons of the model watch here: youtube.com/watch?v=X7gKBGVz… ↓ more resources

0:12
Yann LeCun avatar
⟳ WAC2022 retweeted
Yann LeCun@ylecun·17 Feb 2024

The Diffusion Transformer paper, by my former-FAIR-and-current-NYU colleague @sainingxie and former-Berkeley-student-and-current-OpenAI engineer William Peebles, was rejected from CVR2023 for "lack of novelty", accepted at ICCV2023, and apparently forms the basis for Sora. openaccess.thecvf.com/conten…

Here's my take on the Sora technical report, with a good dose of speculation that could be totally off. First of all, really appreciate the team for sharing helpful insights and design decisions – Sora is incredible and is set to transform the video generation community. What we have learned so far: - Architecture: Sora is built on our diffusion transformer (DiT) model (published in ICCV 2023) — it's a diffusion model with a transformer backbone, in short: DiT = [VAE encoder + ViT + DDPM + VAE decoder]. According to the report, it seems there are not much additional bells and whistles. - "Video compressor network": Looks like it's just a VAE but trained on raw video data. Tokenization probably plays a significant role in getting good temporal consistency. By the way, VAE is a ConvNet, so DiT technically is a hybrid model ;) (1/n)

Post attachment
Danfei Xu avatar
⟳ WAC2022 retweeted
Danfei Xu@danfei_xu·5 Apr 2023

One of the most impressive CV works I've seen recently. Also huge kudos to Meta AI for sticking to open sourcing despite the trend increasingly going towards the opposite direction.

AI at Meta@AIatMeta5 Apr 2023

Today we're releasing the Segment Anything Model (SAM) — a step toward the first foundation model for image segmentation. SAM is capable of one-click segmentation of any object from any photo or video + zero-shot transfer to other segmentation tasks ➡️ https://t.co/qYUoePrWVi

0:16
Chris Paxton avatar
⟳ WAC2022 retweeted
Chris Paxton@chris_j_paxton·17 Mar 2023

Might be the most impressive version of this yet

AK@_akhaliq17 Mar 2023

LERF: Language Embedded Radiance Fields TL;DR: Grounding CLIP vectors volumetrically inside a NeRF allows flexible natural language queries in 3D abs: arxiv.org/abs/2303.09553 project page: lerf.io/

0:33
Tansu Yegen avatar
⟳ WAC2022 retweeted
Tansu Yegen@TansuYegen·13 Mar 2023

If a person in some parts of China crosses the street in the improper location, the facial recognition technology will immediately send them to a public shame board, connect them to their bank card, and deduct the amount of the fine

0:19