FreeMesh: Boosting Mesh Generation with Coordinates Merging
This work addresses a specific bottleneck in mesh generation for researchers and practitioners by providing tools to enhance tokenization efficiency, though it is incremental as it builds upon existing tokenization methods.
The paper tackles the lack of efficient evaluation for mesh tokenizers in auto-regressive mesh generation by introducing a new metric called Per-Token-Mesh-Entropy (PTME) to assess tokenizers without training, and proposes a coordinate merging technique that improves compression ratios by rearranging and merging frequent coordinate patterns, as validated through experiments on methods like MeshXL and MeshAnything V2.
The next-coordinate prediction paradigm has emerged as the de facto standard in current auto-regressive mesh generation methods. Despite their effectiveness, there is no efficient measurement for the various tokenizers that serialize meshes into sequences. In this paper, we introduce a new metric Per-Token-Mesh-Entropy (PTME) to evaluate the existing mesh tokenizers theoretically without any training. Building upon PTME, we propose a plug-and-play tokenization technique called coordinate merging. It further improves the compression ratios of existing tokenizers by rearranging and merging the most frequent patterns of coordinates. Through experiments on various tokenization methods like MeshXL, MeshAnything V2, and Edgerunner, we further validate the performance of our method. We hope that the proposed PTME and coordinate merging can enhance the existing mesh tokenizers and guide the further development of native mesh generation.