Result details - ETH3D

Submitted anonymously.

Submission data

Full name	Depth-SupervisedFeatureMatching
Description	A stereo visual SLAM approach that combines deep feature extraction with geometry-aware matching. The method generates dense depth priors from stereo images and refines feature correspondences using temporal consistency and IMU constraints. A robust pose graph optimization module removes outliers and improves trajectory accuracy in challenging environments with repetitive textures and dynamic objects.
Parameters	Feature points: 2500 Matching confidence threshold: 0.75 Keyframe interval: 8 frames IMU fusion weight: 0.35 Loop closure threshold: 0.82 Bundle adjustment window: 15 keyframes
Programming language(s)	C++, Python
Hardware	A100
Submission creation date	23 Jun, 2026
Last edited	23 Jun, 2026

High-res multi-view results

Info	all	high-res multi-view	indoor	outdoor	courty.	delive.	electro	facade	kicker	meadow	office	pipes	playgr.	relief	relief.	terrace	terrai.
No results yet.

Low-res many-view results

Info	all	low-res many-view	indoor	outdoor	lakeside	sand box	storage room	storage room 2	tunnel
No results yet.

Low-res two-view results

Info	all	lakes. 1l	lakes. 1s	sand box 1l	sand box 1s	stora. room 1l	stora. room 1s	stora. room 2l	stora. room 2s	stora. room 2 1l	stora. room 2 1s	stora. room 2 2l	stora. room 2 2s	stora. room 3l	stora. room 3s	tunnel 1l	tunnel 1s	tunnel 2l	tunnel 2s	tunnel 3l	tunnel 3s
							0.45	8.16	5.19	5.32	13.23	7.80	4.28	13.33	15.72	0.13	0.07	0.00	0.00	0.20

SLAM results

		all	boxes	boxes dark	buddha	cables 4	cables 5	desk 1	desk 2	desk changing 2	desk dark 1	desk dark 2	desk global light changes	desk ir light	dino	drone	foreground occlusion	helmet	kidnap 2	lamp	large loop 2	large loop 3	large non loop	motion 2	motion 3	motion 4	planar 1	reflective 2	scale change	table 1	table 2	table 5	table 6	table global light changes	table local light changes	table scene	trashbin
	Method	Info
No results yet.