This paper presents a fast adaptive image restoration method for removing spatially varying out-of-focus blur of a general imaging sensor. After estimating the parameters of space-variant point-spread-function (PSF) using the derivative in each uniformly blurred region, the proposed method performs spatially adaptive image restoration by selecting the optimal restoration filter according to the estimated blur parameters. Each restoration filter is implemented in the form of a combination of multiple FIR filters, which guarantees the fast image restoration without the need of iterative or recursive processing. Experimental results show that the proposed method outperforms existing space-invariant restoration methods in the sense of both objective and subjective performance measures. The proposed algorithm can be employed to a wide area of image restoration applications, such as mobile imaging devices, robot vision, and satellite image processing.

This paper presents a single fisheye lens camerabased visual surveillance system for monitoring a wide area. A fisheye lens has a wider field-of-view (FOV) than normal lenses at the cost of a barrel distortion in the acquired image. After correcting the barrel distortion, the proposed algorithm detects objects, and performs tracking using a histogram-based Gaussian mixture model (GMM). Experimental results show that the proposed algorithm can efficiently detect objects by reducing the geometric distortion of the input image. For this reason it is suitable for not only surveillance cameras but also consumer applications of video object detection and recognition.

This paper presents a depth-based defocus map estimation method from a single camera with multiple off-axis apertures. The proposed estimation algorithm consists of two steps: (i) object distance estimation using off-axis apertures and (ii) defocus map estimation based on the object distance. The proposed method can accurately estimate the defocus map using object distances that are well-characterized in a color shift model-based computational camera. Experimental results show that the proposed method outperforms the state-of-the-art defocus estimation methods in the sense of both accuracy and the estimation range. The proposed defocus map estimation method is suitable for multifocusing, refocusing, and extended depth of field (EDoF) systems.

This paper presents a robust background generation method using a modified mixture of Gaussian model. Traditional background generation methods become unstable when the camera viewpoint suddenly changes. To solve this problem, the proposed method robustly extracts the background by mixing multiple Gaussian models.

This paper presents a video completion algorithm using block matching for video stabilization. In order to fill in missing pixels, the proposed algorithm consists of three steps: i) mosaicking for covering the missing static, planar regions, ii) estimation of local motion vectors using the hierarchical LucasKanade optical flow method, and iii) selection of the most similar patch in both spatial and temporal neighbors. The proposed video completion algorithm can be applied in the wide areas of consumer electronics including camcorders, smart phone cameras, tablet cameras, and smart glasses.