Nonlinear Depth Quantization Using Piecewise Linear Scaling for Immersive Video Coding

Abstract

Moving Picture Experts Group (MPEG) is developing a standard for immersive video coding called MPEG Immersive Video (MIV) and is releasing a reference software called Test Model for Immersive Video (TMIV) in the standardization process. The TMIV efficiently compresses an immersive video comprising a set of texture and depth views acquired using multiple cameras within a limited 3D viewing space. Moreover, it affords a rendered view of an arbitrary view position and orientation with six degrees of freedom (6DoF). However, the existing depth quantization applied to depth atlas in TMIV is insufficient since the reconstructed depth is crucial for achieving the required quality of a rendered viewport. To address this issue, we propose a nonlinear depth quantization method that allocates more codewords to a depth subrange with a higher occurrence of depth values located at edge regions, which are important in terms of the rendered view quality. We implement the proposed nonlinear quantization based on piecewise linear scaling considering the computational complexity and bitstream overhead. The experimental results show that the proposed method yields PSNR-based Bjøntegaard delta rate gains of 5.2% and 4.9% in the end-toend performance for High-and Low-bitrate (BR) ranges, respectively. Moreover, subjective quality improvement is mainly observed at the object boundaries of the rendered viewport. The proposed nonlinear quantization method has been adopted into the TMIV as a candidate standard technology for the next MIV edition.

A Survey on Video Streaming for Next-Generation Vehicular Networks

“…In contrast, MIV addresses inter-view redundancies during coding, providing improved compression by leveraging geometry information through depth maps. Recent studies [148,149] have focused on enhancing the quality of depth maps in MIV. Lee et al [150] proposed a groupbased adaptive rendering method for volumetric video streaming.…”

AE-Qdrop: Towards Accurate and Efficient Low-Bit Post-Training Quantization for A Convolutional Neural Network

“…Table 1 summarizes some classical post-training quantization schemes. Early PTQ focused on minimizing the quantization error of network parameters through techniques such as optimizing quantization factor scale [21,22], bias correction [27,28], piecewise linear quantization [29,30], and outlier separation [31,32]. For example, Nvidia's TensorRT [22], a widely used quantization tool, searched for the optimal quantization factor scale by minimizing the Kullback-Leibler (KL) distance between FP activation and quantized activation.…”

“…The Quantization Error of Network Parameters Optimizing Quantization Factor Scale [21,22] Bias Correction [27,28] Piecewise Linear Quantization [29,30] Outlier Separation [31,32] ≤4 bit Layer-wise Reconstruction LAPQ [23] AdaRound [24] AdaQuant [35] Block-wise Reconstruction BrecQ [25] RAPQ [33] Mr.BiQ [34] Qdrop [26] 3. Background and Theoretical Analysis 3.1.…”

Advanced HEVC Screen Content Coding for MPEG Immersive Video

“…Several papers regard the atlas preparation step, including the patch merging [12], allowing for more efficient packing of non-pruned information within atlases, non-linear depth quantization [13], or adaptation of patch occupancy [14] to the block-based nature of typical video encoders, such as HEVC or VVC. Other works focus on rendering and increasing the quality of synthesized viewports for immersive video [15] and non-Lambertian content [16].…”