The Constraints between Edge Depth and Uncertainty for Monocular Depth Estimation

The self-supervised monocular depth estimation paradigm has become an important branch of computer vision depth-estimation tasks. However, the depth estimation problem arising from object edge depth pulling or occlusion is still unsolved. The grayscale discontinuity of object edges leads to a relatively high depth uncertainty of pixels in these regions. We improve the geometric edge prediction results by taking uncertainty into account in the depth-estimation task. To this end, we explore how uncertainty affects this task and propose a new self-supervised monocular depth estimation technique based on multi-scale uncertainty. In addition, we introduce a teacher–student architecture in models and investigate the impact of different teacher networks on the depth and uncertainty results. We evaluate the performance of our paradigm in detail on the standard KITTI dataset. The experimental results show that the accuracy of our method increased from 87.7% to 88.2%, the AbsRel error rate decreased from 0.115 to 0.11, the SqRel error rate decreased from 0.903 to 0.822, and the RMSE error rate decreased from 4.863 to 4.686 compared with the benchmark Monodepth2. Our approach has a positive impact on the problem of texture replication or inaccurate object boundaries, producing sharper and smoother depth images.

Efficient Transformer for Remote Sensing Image Segmentation

Semantic segmentation for remote sensing images (RSIs) is widely applied in geological surveys, urban resources management, and disaster monitoring. Recent solutions on remote sensing segmentation tasks are generally addressed by CNN-based models and transformer-based models. In particular, transformer-based architecture generally struggles with two main problems: a high computation load and inaccurate edge classification. Therefore, to overcome these problems, we propose a novel transformer model to realize lightweight edge classification. First, based on a Swin transformer backbone, a pure Efficient transformer with mlphead is proposed to accelerate the inference speed. Moreover, explicit and implicit edge enhancement methods are proposed to cope with object edge problems. The experimental results evaluated on the Potsdam and Vaihingen datasets present that the proposed approach significantly improved the final accuracy, achieving a trade-off between computational complexity (Flops) and accuracy (Efficient-L obtaining 3.23% mIoU improvement on Vaihingen and 2.46% mIoU improvement on Potsdam compared with HRCNet_W48). As a result, it is believed that the proposed Efficient transformer will have an advantage in dealing with remote sensing image segmentation problems.

The Finger-Based Interactive Projection using a Monocular Camera

Ordinary projection screen is not sensitive to interaction, it cannot meet the demands of teaching, virtual reality, and other applications. Due to the fact that people always use hands to complete a variety of human–computer interaction, the finger-based interactive projection technology is worth being researched. In this paper, an ordinary monocular camera is used to acquire video frame on projection screen, and the touch signal of finger in frame is used as the input of interactive projection system. Because the differences between spatial frequency of common digital camera and the projection screen is small, the frame obtained from camera will contain moire fringe, which needs to be filtered in image frequency domain. Then the difference between current frame edge and previous frame edge is calculated to obtain moving object edge clues. According to these clues, the most possible contour curve is searched in current frame edge, and the curve is fitted by polynomial approximation method. Its curvature integration is used to match with the curvature integration of finger template curve. After that the fingers in the curve are recognized. Because color information is not needed, this method can be used to recognize gloved fingers. Finally, finger shadow is used to judge whether the finger touches projection screen to complete interactive process. The experiments of writing and collaboratively rotating picture on projector screen show that this method can effectively complete interactive operation with the projection screen and can realize the multi-user operation.

Detail Preserved Surface Reconstruction from Point Cloud

In this paper, we put forward a new method for surface reconstruction from image-based point clouds. In particular, we introduce a new visibility model for each line of sight to preserve scene details without decreasing the noise filtering ability. To make the proposed method suitable for point clouds with heavy noise, we introduce a new likelihood energy term to the total energy of the binary labeling problem of Delaunay tetrahedra, and we give its s-t graph implementation. Besides, we further improve the performance of the proposed method with the dense visibility technique, which helps to keep the object edge sharp. The experimental result shows that the proposed method rivalled the state-of-the-art methods in terms of accuracy and completeness, and performed better with reference to detail preservation.

Texture Construction Edge Detection Algorithm

The edge detection algorithm is the cornerstone of image processing; a good edge detection result can further extract the required information through rich texture information and achieve object detection, segmentation, and identification. To obtain a rich texture edge detection technology, this paper proposes using edge texture change for edge construction and constructs the edge contour through constructing an edge texture extension between the blocks to reduce the missing edge problem caused by the threshold setting. Finally, through verification of the experimental results, the proposed method can effectively overcome the problem caused by unsuitable threshold setting and detect rich object edge information compared to the adaptive edge detection method.