Nicholas Fry

PhD Candidate in Computing @imperialcollege

[Shinjeong Kim @shinjeong99]

(1/5) Excited to introduce MLP-Splatting, a novel way to decompose a scene into object-/part-level primitives!

Each object can be represented with a few tens of light-field-based primitives (c.f. 3DGS: hundreds+).

w/ @cheng_yuzh21715 @XinKong_IC @paulhjkelly @AjdDavison

[Nicholas Fry @deepfry_n]

@janheld14 Thank you!! Looking forward to the Mesh Splatting presentation :)

[Nicholas Fry @deepfry_n]

(1/3) I am happy to share our work on Triangle Splatting SLAM!

We show the first RGB-D SLAM to use differentiable triangles as a 3D map representation.

Our method enables online mesh-based deformations and collision checking via on-the-fly Delaunay triangulation.

[Nicholas Fry @deepfry_n]

(3/3) Delaunay triangulation allows us to recover thin connected structures like these branches. I find the deformations particularly satisfying.

[Nicholas Fry @deepfry_n]

(2/3) Work with @eric_dexheimer, @makezur, @paulhjkelly, @AjdDavison

https://t.co/QDpEXAW27T

Here are some fun demos!

[Shinjeong Kim @shinjeong99]

Join us on 3 Jun. at 8 AM in Room 4EF for the On-Sensor Vision Workshop at @CVPR! #CVPR2026
https://t.co/QwO0g1sqzQ

Through invited talks and posters, we’ll explore the future of on-sensor vision: low-power/latency edge computer vision enabled by sensor-side computation.