Submitted by Dominik Hirner.

Submission data

Full namefully-convolutional densely-connected neural network
DescriptionWe propose a novel lightweight network for stereo estimation. The method uses densely connected layer structures to learn expressive features without the need of fully-connected layers or 3D convolutions. This leads to a network structure with only 0.37M parameters while still having competitive results. The post-processing consists of filtering, a consistency check and hole filling. This paper has been accepted to the ICPR 2020 conference in Milan which will be held on the 10-15 January 2021. Therefore this work has not yet been presented
Parameters\eta = 6 \times 10^{-6}
Programming language(s)python3, pytorch
HardwareGeForce 2080 GTX
Source code or download URL
Submission creation date25 Mar, 2020
Last edited9 Nov, 2020

High-res multi-view results

multi-view co.statueterrac.
No results yet.

Low-res many-view results

indooroutdoorlakesidesand boxstorage roomstorage room 2tunnel
No results yet.

Low-res two-view results

Infoalllakes. 1llakes. 1ssand box 1lsand box 1sstora. room 1lstora. room 1sstora. room 2lstora. room 2sstora. room 2 1lstora. room 2 1sstora. room 2 2lstora. room 2 2sstora. room 3lstora. room 3stunnel 1ltunnel 1stunnel 2ltunnel 2stunnel 3ltunnel 3s
copylefttwo views10.720.524.271.881.6317.185.2918.2019.6928.5034.5134.0321.4815.8911.

SLAM results

allboxesboxes darkbuddhacables 4cables 5desk 1desk 2desk changing 2desk dark 1desk dark 2desk global light changesdesk ir lightdinodroneforeground occlusionhelmetkidnap 2lamplarge loop 2large loop 3large non loopmotion 2motion 3motion 4planar 1reflective 2scale changetable 1table 2table 5table 6table global light changestable local light changestable scenetrashbin
No results yet.