NLab-UTokyo at ImageCLEF 2013 Plant Identification Task.
Hideki Nakayama · 2013
Abstract. We describe our system at the ImageCLEF 2013 plant identi-fication task. Plant identification is extremely challenging because target classes are often visually quite similar. To distinguish them, we need to extract highly informative visual features. We believe that the key to achieving this is to enhance the discriminative power of local descrip-tors. We employed multiple local features with our polynomial embed-ding technique to boost the performance. Further, they were encoded into the sophisticated Fisher Vector representation which enables accu-rate classification with linear classifiers. Our system achieved promising performance, and got the first place in NaturalBackground and the third place in SheetAsBackground tasks, respectively.