1. compute the mean image.
<pre name="code" class="plain">#!/usr/bin/env sh
# Compute the mean image from the imagenet training leveldb
# N.B. this is available in data/ilsvrc12
./build/tools/compute_image_mean /mnt_data/yilin_guo/data/food_detection/train_data/ \
./examples/food_detection/googleNet/imagenet_mean.binaryproto
echo "Done."
2. how to convert binary proto to npy
<pre name="code" class="python">import caffe
import numpy as np
import sys
if len(sys.argv) != 3:
print "Usage: python convert_protomean.py proto.mean out.npy"
sys.exit()
blob = caffe.proto.caffe_pb2.BlobProto()
data = open( sys.argv[1] , 'rb' ).read()
blob.ParseFromString(data)
arr = np.array( caffe.io.blobproto_to_array(blob) )
out = arr[0]
np.save( sys.argv[2] , out )