yolo2
yolo2 copied to clipboard
这个项目是基于论文YOLO9000: Better, Faster, Stronger的keras(backend:tensorflow)实现
有wordtree部分的实现吗?
#需要和之前的image_tmp_output融合了 def fun(x): import tensorflow as tf return tf.space_to_depth(x, block_size=2) tmp2 = Lambda(fun,output_shape = (13,13,256),name='space_to_depth_2')(tmp2) tmp = merge(n ame='merge_1',mode='concat')([tmp2,tmp]) 这个cell不能运行,错误如下: ValueError: Dimension size must be evenly divisible by 2 but...
intersect_min = np.minimum(box_min, anchors_min) intersect_max = np.maximum(box_max, anchors_max) 这两句中的取极值写反了吧
anchors的精度不太够,导致detect出来的边框比较大,建议换成0.57273, 0.677385, 1.87446, 2.06253, 3.33843, 5.47434, 7.88282, 3.52778, 9.77052, 9.16828会好很多