def train_bow_sift(id_upper): if not id_upper: id_upper = 3900 sift = Sift() kmeans = KmeansModel() image_list = sorted([x for x in os.listdir(DATAPATH) if os.path.splitext(x)[1] == '.jpg' and int(os.path.splitext(x)[0]) < id_upper],key=lambda x: int(os.path.splitext(x)[0])) sift.compute(image_list) if not kmeans.load('kmeans_sift'): kmeans.fit(sift.descriptors_list) kmeans.save('kmeans_sift') bow = Bow(kmeans) # bag of words of samples # label indicator1 indicator2 ... # ... ... ... ... bow.train_sift(sift.descriptors_list)
def test_something(self): id_upper = 3900 sift = Sift() kmeans = KmeansModel() image_list = sorted([x for x in os.listdir(DATAPATH) if os.path.splitext(x)[1] == '.jpg' and int(os.path.splitext(x)[0]) < id_upper],key=lambda x: int(os.path.splitext(x)[0])) print image_list descriptors_list = sift.compute(image_list) kmeans.fit(descriptors_list) kmeans.save('kmeans_sift') print 'ok'
def test_something(self): sift = Sift() des = sift.compute(['3.jpg','4.jpg']) print des