22 September 2025 in IEEE Transactions on Multimedia
Point cloud oversegmentation method can obtain a series of superpoints by grouping points that are semantically and geometrically consistent. The generated superpoints can be treated as the basic processing units in various downstream tasks to improve task performance and processing efficiency. However, due to the high semantic and geometric complexity of point cloud scenes, obtaining high-quality superpoints is still challenging. Aiming to generate high-quality indoor superpoints, we propose an end-to-end supervised contrastive learning framework SCL-OverSeg for indoor point cloud oversegmentation. Firstly, to solve the challenge of balancing the importance of geometric similarity and spatial proximity constraint between points and superpoints in indoor scenes, we integrate the geometric similarity and spatial proximity constraint into the supervision signal by generating the superpoint ground truth. To solve the challenge of superpoints crossing objects, we propose to utilize instance labels rather than semantic labels to generate the ideal superpoint ground truth as the object-level supervision signal. Secondly, to construct the distinguishable embedding space facilitating to the assignments of points to superpoints, we propose point-superpoint contrastive learning to compel the network to project each point to be closer to the reasonable superpoint in embedding space. Besides, with the instance labels, to improve the superpoint performance on object boundaries, we propose the object boundary contrastive learning to enhance the feature distinguishability between tough points across the object boundaries. Extensive experiments demonstrate that SCL-OverSeg can effectively improve indoor oversegmentation performance, especially on object boundaries. The relevant codes will be available on https://github.com/sssssyf/SCL-OverSeg.
16 January 2026 in ISPRS Journal of Photogrammetry and Remote Sensing
6 March 2026 in Pattern Recognition
12 January 2026 in IEEE Geoscience and Remote Sensing Letters
Diffusion-based shadow removal methods have demonstrated impressive results on natural images, but face challenges such as high computational cost and content distortion when dealing with remote sensing images. To address these limitations, a dual-branch cross-frequency interaction diffusion model DCFI-Net for shadow removal in remote sensing images is proposed. In DCFI-Net, the diffusion process is transferred to the wavelet domain to reduce computational cost. Within the wavelet-based diffusion framework, two parallel dedicated branches for the low-frequency (LF) and high-frequency (HF) components are proposed. Then, a cross-frequency interaction module (CFIM) is utilized to facilitate the information exchange between these two branches, which guides the synergistic network optimization. Finally, we propose a feature preserving module (FPM) to mitigate the scene discrepancies between the shadow and shadow-free images. The experimental results conducted on the UAV-SC and SRGTA datasets demonstrate that DCFI-Net significantly outperforms existing state-of-the-art methods in terms of both the reconstruction and perceptual quality. Notably, DCFI-Net exhibits substantial advantages in precisely reconstructing edge details and preserving scene consistency, which are crucial for remote sensing image analysis.
1 January 2026 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
14 May 2026 in ISPRS Journal of Photogrammetry and Remote Sensing