VQV Signal

RESEARCH WATCH TECHNICAL

Apple posts Depth Pro 2 for metric depth from images and video on Hugging Face

According to a post on X by @_akhaliq, Apple has posted Depth Pro 2 on Hugging Face in four sizes, from 24M to 872M parameters, along with a separate focal-length estimator. The model is described as estimating metric depth from single images, video and live camera feeds. Apple claims sharper edges and less frame-to-frame jitter than prior work, but has published few numbers so far. The weights are licensed for non-commercial research only, and the post is mainly a short demo video.

Source: Scobleizer News Pulse (X) · x.com Found via Robert Scoble’s Scobleizer News Pulse · a post on X, not independently verified Published 2026-11-20T02:40:00+00:00 Detected 2026-11-20T10:02:21+00:00
View original source

According to a post on X by @_akhaliq, Apple has posted Depth Pro 2 on Hugging Face in four sizes, from 24M to 872M parameters, along with a separate focal-length estimator. The model is described as estimating metric depth from single images, video and live camera feeds. Apple claims sharper edges and less frame-to-frame jitter than prior work, but has published few numbers so far. The weights are licensed for non-commercial research only, and the post is mainly a short demo video.

AI-assisted summary based on the listed source.

Apple posted Depth Pro 2 on Hugging Face in four sizes from 24M to 872M parameters plus a separate focal-length estimator, for metric depth from single images, video and live camera feeds. Apple claims sharper edges and less frame-to-frame jitter than prior work but has published few numbers so far, and the...

For robotics and computer vision developers, the release could be relevant to depth perception from video and live camera input. The non-commercial license limits use in commercial products for now. Teams interested in building on it should check the license terms and watch for independent benchmark results that test Apple's claims about edge sharpness and frame-to-frame stability.

Signal Strength 91% Technical label WATCH Public Interest 40 Category RESEARCH Reader Depth TECHNICAL

Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.

Public Interest components
Recognizable Entity Score 0 Practical Impact Score 18 Novelty Interest Score 94 Consequence Score 30 Curiosity Score 84 Shareability Score 48

VQV surfaced this signal because it is recent, relevant to Robotics, connected to Scobleizer News Pulse (X).