This is an example XCode project using iOS version of Caffe built by aleph7.
It also includes a class Classifier which is the cpp_classification imported from Caffe C++ example.
It classifies example image as following screenshot:
- opencv2.framework
- Xcode 6 (can't build with Xcode 7)
Clone it:
$ git clone --recursive git@github.com:noradaiko/caffe-ios-sample.git
You need your caffemodel, deploy.prototxt, mean.binaryproto and labels.txt into model/ directory.
This sample already includes files for testing except for caffemodel.
You can download BVLC CaffeNet Model from: http://dl.caffe.berkeleyvision.org/bvlc_reference_caffenet.caffemodel.
Of course, you can use your own network model if you have.
NSString* model_file = [NSBundle.mainBundle pathForResource:@"deploy" ofType:@"prototxt" inDirectory:@"model"];
NSString* label_file = [NSBundle.mainBundle pathForResource:@"labels" ofType:@"txt" inDirectory:@"model"];
NSString* mean_file = [NSBundle.mainBundle pathForResource:@"mean" ofType:@"binaryproto" inDirectory:@"model"];
NSString* trained_file = [NSBundle.mainBundle pathForResource:@"bvlc_reference_caffenet" ofType:@"caffemodel" inDirectory:@"model"];
string model_file_str = std::string([model_file UTF8String]);
string label_file_str = std::string([label_file UTF8String]);
string trained_file_str = std::string([trained_file UTF8String]);
string mean_file_str = std::string([mean_file UTF8String]);
UIImage* example = [UIImage imageNamed:@"image_0002.jpg"];
cv::Mat src_img;
UIImageToMat(example, src_img);
Classifier classifier = Classifier(model_file_str, trained_file_str, mean_file_str, label_file_str);
std::vector<Prediction> result = classifier.Classify(src_img);Output into console:
for (std::vector<Prediction>::iterator it = result.begin(); it != result.end(); ++it) {
NSString* label = [NSString stringWithUTF8String:it->first.c_str()];
NSNumber* probability = [NSNumber numberWithFloat:it->second];
NSLog(@"label: %@, prob: %@", label, probability);
}Caffe is released under the BSD 2-Clause license. The BVLC reference models are released for unrestricted use.
Please cite Caffe in your publications if it helps your research:
@article{jia2014caffe,
Author = {Jia, Yangqing and Shelhamer, Evan and Donahue, Jeff and Karayev, Sergey and Long, Jonathan and Girshick, Ross and Guadarrama, Sergio and Darrell, Trevor},
Journal = {arXiv preprint arXiv:1408.5093},
Title = {Caffe: Convolutional Architecture for Fast Feature Embedding},
Year = {2014}
}