forked from s-gupta/rcnn-depth
-
Notifications
You must be signed in to change notification settings - Fork 0
/
get_paths.m
44 lines (36 loc) · 1.8 KB
/
get_paths.m
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
function p = get_paths(runname)
% function p = get_paths(runname)
% AUTORIGHTS
if ~exist('runname', 'var')
global RUNNAME
if isempty(RUNNAME), runname = 'release'
else, runname = RUNNAME; end
end
p.root_cache_dir = fullfile(pwd(), '..', 'eccv14-cachedir');
p.cache_dir = fullfile(p.root_cache_dir, runname);
p.contours_dir = fullfile(p.cache_dir, 'contours');
p.contours_cues_dir = fullfile(p.contours_dir, 'cues');
p.contours_model_dir = fullfile(p.contours_dir, 'models');
p.output_dir = fullfile(p.cache_dir, 'output');
p.ucm_dir = fullfile(p.output_dir, 'ucm');
p.box_dir = fullfile(p.cache_dir, 'regions');
% p.caffe_dir = '/home/eecs/sgupta/eccv14-code/cachedir/release/detection/';
% p.ft_image_dir = fullfile(p.caffe_dir, 'ft_images');
% p.ft_hha_dir = fullfile(p.caffe_dir, 'ft_hha');
p.detectior_dir = fullfile(p.cache_dir, 'detection');
p.cnnF_cache_dir = fullfile(p.detectior_dir, 'feat_cache');
p.ft_image_dir = fullfile(p.detectior_dir, 'ft_images');
p.ft_hha_dir = fullfile(p.detectior_dir, 'ft_hha');
p.ft_disparity_dir = fullfile(p.detectior_dir, 'ft_disparity');
p.ft_dir = fullfile(p.detectior_dir, 'finetuning');
p.wf_dir = fullfile(p.ft_dir, 'windowfile');
p.proto_dir = fullfile(p.ft_dir, 'protodir');
p.snapshot_dir = fullfile(p.ft_dir, 'snapshot');
f = fieldnames(p);
for i = 1:length(f), exists_or_mkdir(p.(f{i})); end
p.mean_file_color = fullfile('caffe-data', 'mean', 'imagenet_mean');
p.mean_file_disparity = fullfile('caffe-data', 'mean', 'nyu_disparity_train1');
p.mean_file_hha = fullfile('caffe-data', 'mean', 'nyu_hha_train1');
p.caffe_net = fullfile('caffe-data', 'caffe_reference_imagenet_model');
p.rgb_edge_model = fullfile('structured-edges', 'models', 'forest', 'modelBsds.mat');
end