Superpixels: An evaluation of the state-of-the-art

Depth boundaries often lose sharpness when upsampling from low-resolution (LR) depth maps especially at large upscaling factors. We present a new method to address the problem of depth map super resolution in which a high-resolution (HR) depth map is inferred from a LR depth map and an additional HR intensity image of the same scene. We propose a Multi-Scale Guided convolutional network (MSG-Net) for depth map super resolution. MSG-Net complements LR depth features with HR intensity features using a multi-scale fusion strategy. Such a multi-scale guidance allows the network to better adapt for upsampling of both fine- and large-scale structures. Specifically, the rich hierarchical HR intensity features at different levels progressively resolve ambiguity in depth map upsampling. Moreover, we employ a high-frequency domain training method to not only reduce training time but also facilitate the fusion of depth and intensity features. With the multi-scale guidance, MSG-Net achieves state-of-art performance for depth map upsampling.

Depth Map Super-Resolution by Deep Multi-Scale Guidance

This paper proposes an adaptive color-guided autoregressive (AR) model for high quality depth recovery from low quality measurements captured by depth cameras. We observe and verify that the AR model tightly fits depth maps of generic scenes. The depth recovery task is formulated into a minimization of AR prediction errors subject to measurement consistency. The AR predictor for each pixel is constructed according to both the local correlation in the initial depth map and the nonlocal similarity in the accompanied high quality color image. We analyze the stability of our method from a linear system point of view, and design a parameter adaptation scheme to achieve stable and accurate depth recovery. Quantitative and qualitative evaluation compared with ten state-of-the-art schemes show the effectiveness and superiority of our method. Being able to handle various types of depth degradations, the proposed method is versatile for mainstream depth sensors, time-of-flight camera, and Kinect, as demonstrated by experiments on real systems.

Color-Guided Depth Recovery From RGB-D Data Using an Adaptive Autoregressive Model

We propose an algorithm utilizing geodesic distances to upsample a low resolution depth image using a registered high resolution color image. Specifically, it computes depth for each pixel in the high resolution image using geodesic paths to the pixels whose depths are known from the low resolution one. Though this is closely related to the all-pair-shortest-path problem which has O(n2 log n) complexity, we develop a novel approximation algorithm whose complexity grows linearly with the image size and achieve realtime performance. We compare our algorithm with the state of the art on the benchmark dataset and show that our approach provides more accurate depth upsampling with fewer artifacts. In addition, we show that the proposed algorithm is well suited for upsampling depth images using binary edge maps, an important sensor fusion application.

/pdf/joint-geodesic-upsampling-of-depth-images-1swesl49nb.pdf

Joint Geodesic Upsampling of Depth Images

Depth estimation from a single image is a fundamental problem in computer vision. In this paper, we propose a simple yet effective convolutional spatial propagation network (CSPN) to learn the affinity matrix for depth prediction. Specifically, we adopt an efficient linear propagation model, where the propagation is performed with a manner of recurrent convolutional operation, and the affinity among neighboring pixels is learned through a deep convolutional neural network (CNN). We apply the designed CSPN to two depth estimation tasks given a single image: (1) Refine the depth output from existing state-of-the-art (SOTA) methods; (2) Convert sparse depth samples to a dense depth map by embedding the depth samples within the propagation procedure. The second task is inspired by the availability of LiDAR that provides sparse but accurate depth measurements. We experimented the proposed CSPN over the popular NYU v2 [1] and KITTI [2] datasets, where we show that our proposed approach improves not only quality (e.g., 30% more reduction in depth error), but also speed (e.g., 2 to 5\(\times \) faster) of depth maps than previous SOTA methods. The codes of CSPN are available at: https://github.com/XinJCheng/CSPN.

/pdf/depth-estimation-via-affinity-learned-with-convolutional-49ctbnfjl8.pdf

Depth Estimation via Affinity Learned with Convolutional Spatial Propagation Network

Screen content coding (SCC) has evolved into the extension of the High Efficiency Video Coding (HEVC). Low-latency, real-time transport between devices in the form of screen content video is becoming popular in many applications. However, the complexity of encoder is still very high for intra prediction in HEVC-based SCC. This paper proposes a fast intra prediction method based on content property analysis for HEVC-based SCC. First, coding units (CUs) are classified into natural content CUs (NCCUs) and screen content CUs (SCCUs), based on the statistic characteristics of the content. For NCCUs, the newly adopted prediction modes, including intra block copy mode and palette mode are skipped, if the DC or PLANR mode is the best mode, after testing the traditional intra prediction rough modes. In addition, the quadtree partition process is also terminated due to the homogeneous and smooth block usually chooses a large size CU. For SCCUs, a rank-based decision strategy is introduced to terminate the splitting process of current CU. For all CUs, the bit per pixel of current CU is used to make a CU size decision. Meanwhile, the depth information of neighboring CUs and co-located CU are utilized to further improve the performance. Experimental results show that the proposed algorithm can save 44.92% encoding time on average with negligible loss of video quality.

Fast Intra Prediction Based on Content Property Analysis for Low Complexity HEVC-Based Screen Content Coding

This paper proposes an adaptive color-guided auto-regressive (AR) model for high quality depth recovery from low quality measurements captured by depth cameras. We formulate the depth recovery task into a minimization of AR prediction errors subject to measurement consistency. The AR predictor for each pixel is constructed according to both the local correlation in the initial depth map and the nonlocal similarity in the accompanied high quality color image. Experimental results show that our method outperforms existing state-of-the-art schemes, and is versatile for both mainstream depth sensors: ToF camera and Kinect.

/pdf/depth-recovery-using-an-adaptive-color-guided-auto-4fe58xkgy8.pdf

Depth recovery using an adaptive color-guided auto-regressive model

With the development of three-dimensional (3-D) display technologies, 3-D video has attracted more and more interest. Multiview video plus depth (MVD) is one of the most popular representation formats of 3-D video. In MVD coding system, multiview depth video needs to be coded and transmitted in addition to the texture video. This paper presents a novel fast mode decision (FMD) method for odd views in multiview depth video coding. First, the inter-view and inter-component coding correlations are analyzed to provide efficient reference information. Then, with a view to the characteristics of different types of frames, different early termination strategies are proposed. For the nonanchor frame, the early termination criterion is based on the rate-distortion cost information of the even views and the coded block pattern information. For the anchor frame, the criterion is set stricter to maintain the coding accuracy. Experimental results show that the proposed method can reduce 78.07% coding time on average, without significant loss of video quality.

Fast Mode Decision Using Inter-View and Inter-Component Correlations for Multiview Depth Video Coding

The lack of 3-D content has become a bottleneck for the advancement of three-dimensional television (3-DTV), but conventional multicamera arrays for multiview imaging are expensive to setup and cumbersome to use. This paper proposes a lightweight multiview imaging approach with Kinect, a handheld integrated depth-color camera, under the depth-image-based rendering framework. The proposed method consists of two components: depth restoration from noisy and incomplete depth measurements and view synthesis from depth-color pairs. In depth restoration, we propose a moving 2-D polynomial approximation via least squares to suppress quantization errors in the acquired depth values, and propose a progressive edge-guided trilateral filter to fill missing areas of the depth map. Edges extracted from color image are used to predict the locations of depth discontinuities in missing areas and to guide the proposed trilateral filter avoiding filtering across discontinuities. In view synthesis, we propose a low-rank matrix restoration model to inpaint disocclusion regions, fully exploiting the nonlocal correlations in images, and devise an efficient algorithm under the augmented lagrange multiplier (ALM) framework. Disocclusion areas are inpainted progressively from the boundaries of disocclusion with an estimated priority consisting of four terms: warping term, reliability term, texture term, and depth term. Experimental results show that our method restores high quality depth maps even for large missing areas, and synthesizes natural multiview images from restored depth maps. Strong 3-D visual experiences are observed when the synthesized multiview images are shown in two types of stereoscopic displays.

Chunping Hou

Papers

Color-Guided Depth Recovery From RGB-D Data Using an Adaptive Autoregressive Model

Fast Intra Prediction Based on Content Property Analysis for Low Complexity HEVC-Based Screen Content Coding

Depth recovery using an adaptive color-guided auto-regressive model

Fast Mode Decision Using Inter-View and Inter-Component Correlations for Multiview Depth Video Coding

Computational multi-view imaging with kinect