… classical RGB CNN methods by using the depth information provided by the RGB-D cameras…
or image (HHA) or pass from 2D CNN to 3D CNN. This paper proposes a novel and generic …

… to replace basic CNN operators: depth-adapted convolution and depth-adapted average
pooling… on a frontoparallel plane on the 3D camera basis. While the depth value is constant, our …

… was formulated by Wu 36 for RGB-D object recognition. A novel and generic approach
depth-adapted CNN was proposed by Wu 36 for RGB-D cameras, which was employed in our …

… Second, we investigate the depth-adapted offset which … We argue that the basic assumption
of RGB-D fusion is that these … Deep Learning models from CNN to Transformer via different …

… -spectral (MS) camera and an RGB-D camera that are mounted on … YOLOv5 is a popular
convolutional neural network (CNN) for … However, our RGB-D data contains four channels: three …

… to fuse the two streams of RGB-D for intermediate features' … In this work, we aim for RGB-D
saliency detection that is … To this end, we propose a robust RGB-D fusion method that …

… Abstract—RGB-D semantic segmentation can be advanced with … To solve this problem in
the CNN structure, we propose a Pixel … Demonceaux, “Depthadapted cnns for rgb-d semantic …

… • By exploiting the nature of CNN sampling position, we propose … The raw depth maps for
SOD are obtained by camera … gap between the recent RGB-D models and the previous RGB …

… Within the low-level features extracted by the convolutional neural network, we capture the
… from various indoor scenes recorded by RGB-D cameras from the Microsoft Kinect and is an …

… 2) backlight case containing 128 images, in which the camera faces the light source; and (3)
dark case containing 119 images, in which the camera is in dim light conditions. In all experi…