We introduce ConceptVision, a method that aims for high accuracy in categorizing large number of scenes, while keeping the model relatively simpler and efficient for scalability. The proposed method combines the advantages of both low-level representations and high-level semantic categories, and eliminates the distinctions between different levels through the definition of concepts. The proposed framework encodes the perspectives brought through different concepts by considering them in concept groups. Different perspectives are ensembled for the final decision. Extensive experiments are carried out on benchmark datasets to test the effects of different concepts, and methods used to ensemble. Comparisons with state-of-the-art studies show that we can achieve better results with incorporation of concepts in different levels with different perspectives.
We attack the problem of learning face models for public faces from weakly-labelled images collected from web through querying a name. The data is very noisy even after face detection, with several irrelevant faces corresponding to other people. We propose a novel method, Face Association through Model Evolution (FAME), that is able to prune the data in an iterative way, for the face models associated to a name to evolve. The idea is based on capturing discriminativeness and representativeness of each instance and eliminating the outliers. The final models are used to classify faces on novel datasets with possibly different characteristics. On benchmark datasets, our results are comparable to or better than state-of-the-art studies for the task of face identification.
We attack the problem of learning concepts automatically from noisy web image search results. Going beyond low level attributes, such as colour and texture, we explore weakly-labelled datasets for the learning of higher level concepts, such as scene categories. The idea is based on discovering common characteristics shared among subsets of images by posing a method that is able to organise the data while eliminating irrelevant instances. We propose a novel clustering and outlier detection method, namely Rectifying Self Organizing Maps (RSOM). Given an image collection returned for a concept query, RSOM provides clusters pruned from outliers. Each cluster is used to train a model representing a different characteristics of the concept. The proposed method outperforms the state-of-the-art studies on the task of learning low-level concepts, and it is competitive in learning higher level concepts as well. It is capable to work at large scale with no supervision through exploiting the available sources.