daijifeng001 / R-FCN

R-FCN: Object Detection via Region-based Fully Convolutional Networks
Other
1.25k stars 543 forks source link

ask for sharing edge box data #33

Open lld533 opened 8 years ago

lld533 commented 8 years ago

Could you please share some codes and data used to enable edge box proposals in your experiments (the last table in Sect. 4.1 in Page 8 of the NIPS 2016 paper)?

I modified the codes in script_rfcn_VOC0712_ResNet50_OHEM_ss.m, voc0712_trainval_ss.m, and voc2007_test_ss.m by simply substituting "with_selective_search" with "with_edge_box". As there is no "edge_box_data" folder made publicly available, I employed the data provided at https://github.com/gidariss/LocNet. I only have ~11..4% training accuracy ~9% test accuracy in training stage, a pretty large gap with 77.8% reported in your paper.

D-X-Y commented 7 years ago

In LocNet, the edge_box data has the coordinate [y1, x1, y2, x2], but in the ss data used in Fast / Faster RCNN and R-FCN is [x1, y1, x2, y2]. Maybe this difference causes your performance reduce??