Differentiable Parsing and Visual Grounding of Verbal Instructions for Object Placement
share
Grounding spatial relations in natural language for object placing could have ambiguity and compositionality issues. To address the issues, we introduce ParaGon, a PARsing And visual GrOuNding framework for language-conditioned object placement. It parses language instructions into relations between objects and grounds those objects in visual scenes. A particle-based GNN then conducts relational reasoning between grounded objects for placement generation. ParaGon encodes all of those procedures into neural networks for end-to-end training, which avoids annotating parsing and object reference grounding labels. Our approach inherently integrates parsing-based methods into a probabilistic, data-driven framework. It is data-efficient and generalizable for learning compositional instructions, robust to noisy language inputs, and adapts to the uncertainty of ambiguous instructions.
READ FULL TEXT