Taichi Yoshida

Two dimensional non-separable adaptive directional lifting structure of discrete wavelet transform

In this paper, we propose a two dimensional (2D) non-separable adaptive directional lifting (ADL) structure of discrete wavelet transform (DWT) and its image coding application. Conventionally, we have proposed a polyphase representation of a 2D non-separable lifting structure of DWT. We generalize the polyphase representation in this paper and one structure of the class has been proven to reduce errors due to the rounding operations and improve a compatibility of the irreversible 9/7 DWT, compared with a 2D separable lifting structure of DWT. Adding the adaptive directional transforming property to the generalized structure, our proposed method improves the lossy image coding performance with maintaining the compatibility.

Bit-depth scalable lossless coding for high dynamic range images

In this paper, we propose a bit-depth scalable lossless coding method for high dynamic range (HDR) images based on a reversible logarithmic mapping. HDR images are generally expressed as floating-point data, such as in the OpenEXR or RGBE formats. Our bit-depth scalable coding approach outputs base layer data and enhancement layer data. It can reconstruct the low dynamic range (LDR) image from the base layer data and reconstructs the HDR image by adding the enhancement layer data. Most previous two-layer methods have focused on the lossy coding of HDR images. Unfortunately, the extension of previous lossy methods to lossless coding does not significantly compress the enhancement layer data. This is because the bit depth becomes very large, especially for HDR images in floating-point data format. To tackle this problem, we apply a reversible logarithmic mapping to the input HDR data. Moreover, we introduce a format conversion to avoid any degradation in the quality of the reconstructed LDR image. The proposed method is effective for both OpenEXR and RGBE formats. Through a series of experiments, we confirm that the proposed method decreases the volume of compressed data while maintaining the visual quality of the reconstructed LDR images.

Rate-dependent seam carving and its application to content-aware image coding

Content-aware image resizing (CAIR) is desired because it preserves prominent regions in a resized image. However, CAIR requires high computational complexity to perform in mobile devices, though it is desired for these displays. Moreover, transmitting the side information for CAIR from the encoder is a problem since it usually requires high bitrates compared with those for image data. In this paper, we present a rate-dependent CAIR method that produce various retargeting results based on the bitrates for side information. Furthermore, we apply the proposed technique to wavelet-based image coding. Our proposed content-aware image coding method provides good performances for both CAIR and image coding.

Range reduction of HDR images for backward compatibility with LDR image processing

This paper proposes a new range reduction method with the minimum amount of quantization error in the L2 norm under the L infinity norm constraint. It is necessary to reduce dynamic range of pixel values of high dynamic range (HDR) images to have backward compatibility with low dynamic range image processing systems. The simplest approach is to truncate lower bit planes in binary representation of pixel values. However it does not have fine granularity of the reduced range, and also it does not utilize the histogram sparseness. Furthermore, it generates significant amount of quantization errors. In this paper, we propose a new range reduction method which can 1) utilize the histogram sparseness, and also 2) minimize variance of the error 3) under a specified maximum absolute value of the error.

Blocking artifacts reduction of DCT compressed image based on block wiener filtering

In this paper, a novel de-blocking algorithm is proposed, using block wiener filtering, shift and weighted average operations. The dissipation of block boundary occur by a combination of the shift and average operations. However, visually annoying artifacts are noticeable around the edge region in highly compressed images. In the proposed method, the wiener attenuation coefficients are given based on the estimated error in a block unit. The DCT encoded images have a large margin of error from one block to another. Hence texture information losses are prevented while smoothening effect is applied to the region around the edge. In examples, the proposed algorithm shows better performance than the previous one.

Wavelet-based content-aware image coding with rate-dependent seam carving

This paper proposes an image coding method incorporated with content-aware image resizing. It is a promising application of our previously proposed rate-dependent seam carving (RD-SC). RD-SC is combined with lifting-based discrete wavelet transform to construct a new multiresolution image representation. Furthermore, the modified parent-children relationship of SPIHT is shown. Our proposed image coding presents comparable performances to the conventional SPIHT even for the original size, and retargeted image quality is satisfactory.

A simplified lattice structure of two-dimensional generalized lapped orthogonal transform (2-D GenLOT) for image coding

In this paper, we propose a lattice structure of two-dimensional (2-D) linear-phase paraunitary filter banks (LPPUFBs) called as 2-D GenLOT. Muramatsu et al. have proposed a lattice structure of 2-D LPPUFBs conventionally. Our 2-D GenLOTs save design parameters and computational costs compared to the conventional 2-D LPPUFBs. We impose the regularity condition on the proposed FBs for image coding application. The 2-D GenLOTs perform an efficient image coding application.

Near lossless coding of sparse histogram images based on zero-skip quantization

This paper introduces a zero-skip quantization (ZS.Q) scheme for the near lossless coding of sparse histogram images. Increases in the range of pixel values and various tone mapping operations on those pixel values mean that the histogram bins often contain no pixels. Recently, this sparseness of the histogram was used to increase the lossless coding performance by introducing histogram packing. This approach was extended to lossy coding by combining spatial quantization and lossless coding. However, such methods do not satisfy the near lossless (NL) condition. In contrast, conventional NL coding such as the JPEG-LS standard satisfies the NL condition, but does not use the histogram sparseness. In this paper, a simple ZS.Q procedure is introduced that uses the histogram sparseness to increase coding efficiency under the NL condition. The proposed method has the following advantages: 1) It guarantees the maximum quantization error is less than some threshold; 2) It can be combined with an arbitrary lossless encoder, such as lossless JPEG 2000 or lossless JPEG-LS; 3) Coding errors do not accumulate under repeat encoding and decoding.

L2 norm optimization of tone mapping for two layer lossless coding of HDR images

This paper optimizes a tone mapping function in a backward compatible lossless coding of high dynamic range (HDR) images. The mapping function is designed so that it maximizes quality of decoded low dynamic range (LDR) images. An optimization procedure for a lossy coding system is extended to a lossless system under the condition that a user can specify arbitrary tone mapping function to produce an LDR image from the original HDR image. Superiority of the proposed method on the rate distortion curves of the LDR image was confirmed to be up to 7.5 dB in PSNR improvement for an image sample.

Non-separable three dimensional discrete wavelet transform with adaptive directional prediction

In this report, we introduce a directionally adaptive three dimensional (3D) prediction to the non-separable 3D discrete wavelet transform (DWT). The proposed 3D DWT has less lifting steps and better prediction performance comparing to the existing separable 3D DWT with fixed filter coefficients. The 3D DWT can be applied to compress data volume of volumetric medical images or motion pictures. It is also applicable to analyze such kind of multi-dimensional signals. The proposed method has compatibility with the conventional DWT defined by the JPEG 2000 international standard.