iiinova — where your digital self keeps evolving

Finlay Hudson

@fhudson.bsky.social

Computer Vision PhD student at the University of York. Focusing on Object Permance, Amodal completion and just generally getting computer vision to understand beyond visual features.

Posts Replies Media

Finlay Hudson @fhudson.bsky.social · 03/12/2025

Our research aims to advance systems capable of understanding and explaining the world, tackling tasks like object permanence and structural consistency using human-like context and cues. We would love to chat about this! Come see us at poster session 5 (Friday 11:00-14:00)! #NeurIPS 5/5

020

Finlay Hudson @fhudson.bsky.social · 03/12/2025

Building upon this benchmark, we also introduce a large-scale testing dataset of 10,000 scenes alongside a benchmark method. This is a collaborative work from the University of York between myself, @jadgardner.bsky.social and @willsmithvision.bsky.social 4/5

120

Finlay Hudson @fhudson.bsky.social · 03/12/2025

We introduce TAPVid-360; given query points as coordinates in the first frame, the goal is to track the 3D direction (in the camera coordinate frame) to the scene point. We aim for a model to predict in which relative direction a point is but, unlike 3D, not its distance. 3/5

120

Finlay Hudson @fhudson.bsky.social · 03/12/2025

Humans can create internal world models that complete the full sphere of surrounding information, even when only a fraction is currently visible. However, current AI vision systems lack this, often using an egocentric, frame-by-frame approach with poor memory for what is not immediately visible. 2/5

120

Finlay Hudson @fhudson.bsky.social · 03/12/2025

TAPVid-360 will be at NeurIPS25 this week in sunny San Diego!! This work involves models having to understand beyond a camera’s field of view without the need for expensive 3D data. 👇 1/5

151