CLIP-Driven Open-Vocabulary 3D Scene Graph Generation via Cross-Modality Contrastive Learning

no code implementations • CVPR 2024 • Lianggangxu Chen, Xuejiao Wang, Jiale Lu, Shaohui Lin, Changbo Wang, Gaoqi He

Then to align the image with 3DSG the camera view is treated as a positive sample and other views as negatives.

3d scene graph generation Contrastive Learning +2