Abstract: In the field of computer vision, image feature extraction is the bridge connecting low-level image data with high-level visual tasks such as image matching and object recognition. The SIFT ...
Abstract: Recent open-world representation learning approaches have leveraged CLIP to enable zero-shot 3D object recognition. However, performance on real point clouds with occlusions still falls ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results