SimonSchwaiger/resireg_mini
0.3B • Updated • 100 • 7
Spatially Consistent Vision-Language Models (ReSiReg)
ReSiReg improves spatial consistency across various VLM with dense language grounding. Preprint: https://arxiv.org/abs/2606.19088
-
-
ReSiReg: Towards Spatially Consistent Semantics in Language-Conditioned Robotic Tasks
Paper • 2606.19088 • Published • 9 -
ReSiReg Playground
🗺2Common Robotics Tasks with ReSiReg-Mini
-
ReSiReg Mini
🤖2Dense image-text similarity with ReSiReg Mini
-
DavidSeyserHF/resireg-mini-onnx
Robotics • Updated • 1