9 accepted papers
Existing paradigms for remote sensing change detection are caught in a trade-off: CNNs excel at efficiency but lack global context, while Transformers capture long-range dependencies at a prohibitive computational cost. This paper introduces ChangeRWKV, a new architecture that reconciles this confli
The problem of localization on a large-scale satellite image given a frame of query ground view point clouds remains challenging. Existing LiDAR-to-image cross-view localization methods struggle in large-scale scenarios due to limited semantic alignment and the modality gap between point clouds and …
Open-Vocabulary Aerial Detection (OVAD) and Remote Sensing Visual Grounding (RSVG) have emerged as two key paradigms for aerial scene understanding. However, each paradigm suffers from inherent limitations when operating in isolation: OVAD is restricted to coarse category-level semantics, while RSVG…
This paper introduces BEVPointNet3D, an innovative 3D lane detection model that effectively integrates Bird’s Eye View (BEV) and point cloud features. The proposed approach addresses the inherent limitations of conventional methods that predominantly rely on the flat-ground assumption. BEVPointNet3D
This paper presents a novel feature alignment strategy for cross-view geo-localization to bridge the perspective gap between ground and satellite images. Existing methods for cross-view geo-localization often overlook factors such as occlusion and distortion errors caused by viewpoint transformation
The fusion of infrared and visible images is hard due to their different modalities. Different from existing methods using the integer-order gradient, we design an optimization model to fuse infrared and visible images using fractional-order gradient information. In this way, the complementary infor…
As a voxel-wise labeling task, semantic scene completion (SSC) tries to simultaneously infer the occupancy and semantic labels for a scene from a single depth and/or RGB image. The key challenge for SSC is how to effectively take advantage of the 3D context to model various objects or stuffs with se…
Semantic scene completion (SSC) refers to the task of inferring the 3D semantic segmentation of a scene while simultaneously completing the 3D shapes. We propose PALNet, a novel hybrid network for SSC based on single depth. PALNet utilizes a two-stream network to extract both 2D and 3D features from
RGB images differentiate from depth as they carry more details about the color and texture information, which can be utilized as a vital complement to depth for boosting the performance of 3D semantic scene completion (SSC). SSC is composed of 3D shape completion (SC) and semantic scene labeling whi…