cultivated
发表于 2025-3-26 23:13:12
http://reply.papertrans.cn/47/4615/461484/461484_31.png
endocardium
发表于 2025-3-27 04:36:52
http://reply.papertrans.cn/47/4615/461484/461484_32.png
pantomime
发表于 2025-3-27 07:42:53
Moving Object Detection Based on Self-adaptive Contour Extractiononal neural network to extract more salient features and use region proposal network to generate candidate regions. Afterwards, the feature maps and proposal regions are inputed to ROI pooling layer followed with some fully connected layers to classify objects and regress bounding box. Simulation ex
Insubordinate
发表于 2025-3-27 09:30:29
http://reply.papertrans.cn/47/4615/461484/461484_34.png
赤字
发表于 2025-3-27 17:22:25
Boundary Information Aggregation and Adaptive Keypoint Combination Enhanced Object Detectionoint-based object detection method in order to better locate center keypoints of objects and adaptively combine keypoints to obtain more accurate bounding boxes. Specifically, to better locate center keypoints of objects, we aggregate boundary information by adding the center pooling operation to th
迷住
发表于 2025-3-27 18:29:13
Accurate Oriented Instance Segmentation in Aerial Imagesin aerial images, methods detecting objects with axis-aligned boxes are unsuitable, since the orientation of objects is arbitrary. What’s more, the RoI pooling step existed in these systems results in the loss of spatial details due to the feature warping and resizing, which will degrade the segment
conscribe
发表于 2025-3-28 00:27:19
http://reply.papertrans.cn/47/4615/461484/461484_37.png
高谈阔论
发表于 2025-3-28 03:41:40
Skeleton-Aware Network for Aircraft Landmark Detection problem: .. We have a key observation: the aircraft is a rigid object with global structural relationships between local landmarks. This motivates us to progressively learn the global geometrical structure and local landmark localization in a coarse-to-fine guidance manner. In this paper, we propos
事与愿违
发表于 2025-3-28 09:14:39
Efficient Spectral Pyramid and Spectral-Spatial Feature Interactive Hyperspectral Image Classificatiper, we propose a novel HSI classification method with a deeper network and fewer parameters. Two novel modules named the efficient spectral pyramid (ESP), and improved spectral-spatial feature interactive (SSI) are designed to improving the SS3FCN, which is proposed in our previous work. Specifical
gorgeous
发表于 2025-3-28 13:56:29
http://reply.papertrans.cn/47/4615/461484/461484_40.png