Open jonghoon-park opened 4 years ago
Hello, can you share with me the code to detect image?
Hello, can you share with me the code to detect image?
Hello @ambr89 which do you want code and images? i use DW2TF recent version of github
Thanks
Hi @jonghoon-park with DW2TF you can obtain yolov3-tiny.pb yolov3-tiny.ckpt.meta yolov3-tiny.ckpt.meta yolov3-tiny.ckpt.data-00000-of-00001
How you can test this new weight?
@ambr89 you can download https://drive.google.com/open?id=1ISfbP7F9h0LlHRJ3qK-kTsyU5xkeYF0Q
tf.placeholder argument will be change 'width, height' to 'height, width' in source code def cfg_net(B, H, W, C, net, param, weights_walker, stack, output_index, scope, training, const_inits, verbose): width = int(param["width"]) height = int(param["height"]) channels = int(param["channels"]) net = tf.placeholder(tf.float32, [None, height, width, channels], name=scope) return net
I had change input image size in .cfg file. To convert .meta, and .pb file is fine.
but run the detection, it cannot detect any object. The YoloV3 case: image size : 800x608 is cannot detected anything. image size : 608x608 is fine. The YoloV3-Tiny case: image size : 640x480 is cannot detected anything. image size : 608x608 is fine. image size : 416x416 is fine.
I want to change input image size. Do i something wrong? Plz Let me know. Thanks