Abstract:Continuous-wave Time-of-flight (TOF) range imaging has become a commercially viable technology with many applications in computer vision and graphics. However, the depth images obtained from TOF cameras contain scene dependent errors due to multipath interference (MPI). Specifically, MPI occurs when multiple optical reflections return to a single spatial location on the imaging sensor. Many prior approaches to rectifying MPI rely on sparsity in optical reflections, which is an extreme simplification. In this p… Show more
“…Depth refinement for indoor environment. In the indoor environment, the quality of the depth from commodity RGB-D sensors is not ideal due to the limitation of the sensing technologies [3,37,11]. A lot of works have been proposed to improve the depth using an aligned highresolution color image.…”
In this paper, we propose a deep learning architecture that produces accurate dense depth for the outdoor scene from a single color image and a sparse depth. Inspired by the indoor depth completion, our network estimates surface normals as the intermediate representation to produce dense depth, and can be trained end-to-end. With a modified encoder-decoder structure, our network effectively fuses the dense color image and the sparse LiDAR depth. To address outdoor specific challenges, our network predicts a confidence mask to handle mixed LiDAR signals near foreground boundaries due to occlusion, and combines estimates from the color image and surface normals with learned attention maps to improve the depth accuracy especially for distant areas. Extensive experiments demonstrate that our model improves upon the state-of-the-art performance on KITTI depth completion benchmark. Ablation study shows the positive impact of each model components to the final performance, and comprehensive analysis shows that our model generalizes well to the input with higher sparsity or from indoor scenes. * indicates equal contributions.† indicates corresponding author.
Color Image Sparse Data from LiDARDeepLiDAR: Our Dense Prediction (colored with input color image)DeepLiDAR: Our Dense Prediction (colored with surface normal)
“…Depth refinement for indoor environment. In the indoor environment, the quality of the depth from commodity RGB-D sensors is not ideal due to the limitation of the sensing technologies [3,37,11]. A lot of works have been proposed to improve the depth using an aligned highresolution color image.…”
In this paper, we propose a deep learning architecture that produces accurate dense depth for the outdoor scene from a single color image and a sparse depth. Inspired by the indoor depth completion, our network estimates surface normals as the intermediate representation to produce dense depth, and can be trained end-to-end. With a modified encoder-decoder structure, our network effectively fuses the dense color image and the sparse LiDAR depth. To address outdoor specific challenges, our network predicts a confidence mask to handle mixed LiDAR signals near foreground boundaries due to occlusion, and combines estimates from the color image and surface normals with learned attention maps to improve the depth accuracy especially for distant areas. Extensive experiments demonstrate that our model improves upon the state-of-the-art performance on KITTI depth completion benchmark. Ablation study shows the positive impact of each model components to the final performance, and comprehensive analysis shows that our model generalizes well to the input with higher sparsity or from indoor scenes. * indicates equal contributions.† indicates corresponding author.
Color Image Sparse Data from LiDARDeepLiDAR: Our Dense Prediction (colored with input color image)DeepLiDAR: Our Dense Prediction (colored with surface normal)
“…The proposed technique of polarization enhancement drastically outperforms a state-of-the-art technique for multipath correction, while using fewer images [29]. Refer to the caption of Fig.…”
Section: Robustness In the Wildmentioning
confidence: 96%
“…The coarse depth map contains quantization errors and noise, so Figure 3. A commonly used benchmark scene [13,29]. Combining polarization with Kinect results in improved performance.…”
Section: Correcting Normals From Polarizationmentioning
confidence: 99%
“…Robustness to diffuse multipath: Diffuse multipath has been an active challenge in the ToF community [35,13,29]. The proposed technique of polarization enhancement drastically outperforms a state-of-the-art technique for multipath correction, while using fewer images [29].…”
Coarse depth maps can be enhanced by using the shape information from polarization cues. We propose a framework to combine surface normals from polarization (hereafter polarization normals) with an aligned depth map. Polarization normals have not been used for depth enhancement before. This is because polarization normals suffer from physics-based artifacts, such as azimuthal ambiguity, refractive distortion and fronto-parallel signal degradation. We propose a framework to overcome these key challenges, allowing the benefits of polarization to be used to enhance depth maps. Our results demonstrate improvement with respect to state-of-the-art 3D reconstruction techniques.
“…This aspect allows our method to overcome some of the critical practical limitations of the related imaging methods. The proposed method requires hardware (mostly consumer-grade electronics) that is far less expensive than that required for the TOF-based NLOS imaging [16,24,25,26,27,33,37,39,40], and the method is more robust than the memoryeffect based imaging techniques that have a limited field-ofview [11,21]. Moreover, a recent publication [31] also uses only ordinal digital cameras but would require very specific scene setup (an accidental occlusion) to obtain better performance.…”
Visual object recognition under situations in which the direct line-of-sight is blocked, such as when it is occluded around the corner, is of practical importance in a wide range of applications. With coherent illumination, the light scattered from diffusive walls forms speckle patterns that contain information of the hidden object. It is possible to realize non-line-of-sight (NLOS) recognition with these speckle patterns. We introduce a novel approach based on speckle pattern recognition with deep neural network, which is simpler and more robust than other NLOS recognition methods. Simulations and experiments are performed to verify the feasibility and performance of this approach.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.