S23DR 2024 Challenge
The Problem:
Structured 3D from Images
Produce semantically meaningful measurements or a representation from which we can easily read off the measurements we care about from the ground-level imagery
We care about results, but also privacy
This year we didn’t come up with a solution
how we can safely release the photos,
so made up with derivative modalities
The dataset: 5k scenes, 15k frames, but no RGB images
Inputs:
Output:
CVPR 2024 Competition & HOVER workshop
Prizes
Timeline
Winners
Discretionary prizes
S23DR team
Jack Langerman
Dmytro Mishkin
Ilke Demir
Caner Korkmaz
Daoyi Gao
Hanzhi Chen
Tolga Birdal