Naoto Katsumata

Database retrieval for similar images using ICA and PCA bases

Similar-image retrieval systems are newly presented and examined. The systems use ICA bases (independent component analysis bases) or PCA bases (principal component analysis bases). These bases can contain source images information, however, the indeterminacy of ordering and amplitude on the bases exists due to the PCA and ICA problem formulation per se. But, this paper successfully avoids this difficulty by using weighted inner products of similar bases. A set of opinion test is carried out on 18 systems according to the combination of {similarity measures (ICA, PCA, color histogram), color spaces (RGB, YIQ, HSV), filtering (with, without)}. The color histogram method is a traditional method. The opinion test shows that the presented method of {ICA, HSV, without filtering} is the best. Runners-up are {ICA, HSV or RGB or YIQ, with filtering}. The traditional method is judged to be much inferior. Thus, this papers method is found quite effective to the similar-image retrieval from large databases.

Independent component analysis minimizing convex divergence

A new class of learning algorithms for independent component analysis (ICA) is presented. Starting from theoretical discussions on convex divergence, this information measure is minimized to derive new ICA algorithms. Since the convex divergence includes logarithmic information measures as special cases, the presented method comprises faster algorithms than existing logarithmic ones. Another important feature of this papers ICA algorithm is to accept supervisory information. This ability is utilized to reduce the permutation indeterminacy which is inherent in usual ICA. By this method, the most important activation pattern can be found as the top one. The total algorithm is tested through applications to brain map distillation from functional MRI data. The derived algorithm is faster than logarithmic ones with little additional memory requirement, and can find task related brain maps successfully via conventional personal computer.

Towards the Unification of Human Movement, Animation and Humanoid in the Network

A network environment that unifies the human movement, animation and humanoid is generated. Since the degrees of freedom are different among these entities, raw human movements are recognized and labeled using the hidden Markov model. This is a class of gesture recognition which extracts necessary information transmitted to the animation software and to the humanoid. The total environment enables the surrogate of the human movement by the animation character and the humanoid. Thus, the humanoid can work as a moving computer acting as a remotely located human in the ubiquitous computing environment.

Retrieval-Aware image compression, its format and viewer based upon learned bases

A retrieval-aware image format (rim format) is developed for the usage in the similar-image retrieval. The format is based on PCA and ICA which can compress source images with an equivalent or often better rate-distortion than JPEG. Besides the data compression, the learned PCA/ICA bases are utilized in the similar-image retrieval since they reflect each source images local patterns. Following the format presentation, an image search viewer for network environments (Wisvi; Waseda image search viewer) is presented. Therein, each query is an image per se. The Wisvi system based on the “rim” method successfully finds similar-images from non-uniform network environments. Experiments support that the PCA/ICA methods are viable to the joint compression and retrieval of digital images. Interested test users can download a β-version of the tool for the joint image compression and retrieval from a web site specified in this paper.

Optimization transfer for computational learning: a hierarchy from f-ICA and alpha-EM to their offsprings

Likelihood optimization methods for learning algorithms are generalized and faster algorithms are provided. The idea is to transfer the optimization to a general class of convex divergences between two probability density functions. The first part explains why such optimization transfer is significant. The second part contains derivation of the generalized independent component analysis (ICA). Experiments on brain fMRI maps are reported. The third part discusses this optimization transfer in the generalized expectation-maximization (EM) algorithm. Hierarchical descendants to this algorithm, such as vector quantization and self-organization, are also explained.

Image-to-Image Retrieval Using Computationally Learned Bases and Color Information

New methods for joint compression and Image-to-image retrieval (12I retrieval) are presented. The novelty exists in the usage of computationally learned image bases besides color distributions. The bases are obtained by the Principal Component Analysis and/or the Independent Component Analysis. On the image compression, PCA and ICA outperform the JPEGs DCT This superiority holds even if the bases and superposition coefficients are quantized and encoded. On the 12I retrieval, the precision-recall curve is used to measure the performance. It is found that adding the basis information always increases the baseline ability of the color information. Besides the retrieval evaluation, a unified image format called RIM (Retrieval-aware IMage format) for effective packing of codewords including bases is specified. Furthermore, an image search viewer called Wisvi (Waseda Image Search Viewer) is developed and exploited. A beta-version of all source codes can be down-loaded from a web site given in the text.