MODEL_ZOO.md

Dense Matching Model Zoo

Model	Pre-trained model type	Paper	Description	Link all
PWarpCSFNet_WS	pfpascal / spair	[6]	weakly-supervised	pfpascal / spair
PWarpCSFNet_SS	pfpascal / spair	[6]	strongly-supervised	pfpascal / spair
PWarpCNCNet_WS	pfpascal / spair	[6]	weakly-supervised	pfpascal / spair
PWarpCNCNet_SS	pfpascal / spair	[6]	strongly-supervised	pfpascal / spair
PWarpCCATs_ft_features_SS	pfpascal / spair	[6]	strongly-supervised	pfpascal /
----	--------	-------	---	---
PDCNet_plus	megadepth	[3], [5]	PDC-Net+	model
----	--------	-------	---	---
WarpCSemanticGLUNet	spair	[4]	Original SemanticGLU-Net is finetuned using our warp consistency objective	model
WarpCSemanticGLUNet	pfpascal	[4]	Original SemanticGLU-Net is finetuned using our warp consistency objective	model
SemanticGLUNet	pfpascal	[4]	Original SemanticGLU-Net is finetuned using warp supervision	model
WarpCRANSACFlow	megadepth	[4]		model
WarpCGLUNet	megadepth / megadepth_stage1	[4]		megadepth / megadepth_stage1
GLUNet_star	megadepth / megadepth_stage1	[4]	Baseline for WarpCGLU-Net, trained with warp-supervision loss only	megadepth / megadepth_stage1
----	--------	-------	---	---
PDCNet	megadepth	[3]		model
GLUNet_GOCor_star	megadepth	[3]	corresponds to GLU-Net-GOCor* in PDCNet	model
----	--------	-------	---	---
GLUNet_GOCor	dynamic	[2]		model
GLUNet_GOCor	static	[2]		model
PWCNet_GOCor	chairs_things_ft_sintel	[2]		model
PWCNet_GOCor	chairs_things	[2]		model
GLUNet	dynamic	[2]		model
----	--------	-------	---	---
GLUNet	static (CityScape-DPED-ADE)	[1]		model
SemanticGLUNet	static (CityScape-DPED-ADE)	[1]		model

To download all of them, run the command bash assets/download_pre_trained_models.sh.

All networks are created in 'model_selection.py'. Weights should be put in pre_trained_models/

Evaluation of WarpCRANSACFlow:

The pre-trained weights can directly be used in the RANSAC-Flow repo.