Reproduce “P$^3$-VLM: A Point-based Alternative for Grounded 3D Vision-Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Anna-Maria Halacheva, Jan-Nico Zaech, Sombit Dey, Luc Van Gool, Danda Pani Paudel
P³-VLM uses a single 3D point to query objects in Gaussian-splat scenes, combining location-, task-, and context-aware visual pathways. This design avoids giving the model a box’s dimensions and shape, while retaining scene features. On Nr3D, it achieves 52.1% class accuracy, versus 33.1% for point-adapted LL3DA. Its box variant scores 52.4%, and point coordinates can still carry category information.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.