I am using the implementation of TensorFlow (1.2, GPU) Faster R-CNN ( here ). When I run the VGG16 network (trained by PASCAL VOC07) on some sample images using the demo.py code, the operating time is strange: it changes a lot, even if I supply the same images.
Lateral note: all input images are changed so that the shorter side is 600 pixels, and the longer side is not more than 1000 pixels; therefore, all input images are approximately the same size.
GPU: GTX 1060 6GB
Using tfconfig.gpu_options.allow_growth = True
Image Resolution, 000456.jpg: 500x375; 000188.png: 1224x370
Demo for data / demo / 000456.jpg Detection accepted 2.187s for 300 offers of objects
Demo for data / demo / 000456.jpg Detection accepted 0.141s for 300 offers of objects
Demo for data / demo / 000188.png Detection accepted 1.321s for 300 offers
demo objects for data / demo / 000188.png discovery took 0.112s for 300 object offers
Reorder images:
Demo for data / demo / 000188.png Detection accepted 1.760s for 300 offers of objects
Demo for data / demo / 000188.png Detection accepted 0.109s for 300 offers of objects
Demo for data / demo / 000456.jpg Detection accepted 1.775s for 300 offers
demo objects for data / demo / 000456.jpg The discovery took 0.142s for 300 object offers
tfconfig.gpu_options.per_process_gpu_memory_fraction = 0.8
//000188.png
1.699s 300
//000188.png
0.114s 300
/demo/ 000456.jpg
1.782s 300
/demo/ 000456.jpg
0.142s 300
, , , , ( 0,1 ). .
.
, - / , ?
, , .. (, ..), / (I ).
: PASCAL VOC 2007 ( tfconfig.gpu_options.allow_growth = True). 2,3 0,25 , .
im_detect: 1/4952 2,326s 0,001s
im_detect: 2/4952 2.153s 0.001s
im_detect: 3/4952 2.047s 0.001s
im_detect: 4/4952 1,971s 0,001s
im_detect: 5/4952 1,606s 0,001s
im_detect: 6/4952 1,363s 0,001s
im_detect: 7/4952 1.433s 0,001s
im_detect: 8/4952 1,507s 0,001s
im_detect: 9/4952 1,356s 0,001s
im_detect: 10/4952 1.424s 0,001s
im_detect: 11/4952 1.307s 0,001s
im_detect: 12/4952 1.210s 0.001s
im_detect: 13/4952 1,198s 0,001s
im_detect: 14/4952 1.123s 0,001s
...
im_detect: 60/4952 0,733s 0,001s
im_detect: 61/4952 0,723s 0,001s
im_detect: 62/4952 0,713s 0,001s
im_detect: 63/4952 0,704s 0,001s
im_detect: 64/4952 0,696s 0,001s
im_detect: 65/4952 0.687s 0.001s
...
im_detect: 250/4952 0,463s 0,001s
im_detect: 251/4952 0,462s 0,001s
im_detect: 252/4952 0,468s 0,001s
im_detect: 253/4952 0,466s 0,001s
im_detect: 254/4952 0,470s 0,001s
im_detect: 255/4952 0,469s 0,001s
...
im_detect: 4948/4952 0.245s 0.001s
im_detect: 4949/4952 0,245s 0,001s
im_detect: 4950/4952 0,245s 0,001s
im_detect: 4951/4952 0,245s 0,001s
im_detect: 4952/4952 0,245 0,001