ProGuT：面向森林场景的标签高效全景分割

中文摘要 AI 辅助

森林环境中的全景分割瓶颈不在于语义质量，而在于实例分离；现有的无监督全景方法能生成可用的“stuff”地图，但“thing”质量近乎为零。基于深度或光流的实例发现方法需要并非总是可用的传感器。我们提出ProGuT（原型引导训练），该方法无需每张图像的训练掩码即可生成全景伪标签，仅需未标记图像和一次性的聚类到类别映射。ProGuT对CLIP补丁特征进行聚类，然后通过多尺度几何先验恢复树干实例，该先验利用结构张量来排除非树干结构。与基于深度、光流或类别监督的方法相比，生成伪标签的成本更低。这些伪标签随后用于下游任务，我们将其与其他无监督基线进行了评估。ProGuT在我们的森林数据集上达到了65.2的全景质量（PQ），比初始伪标签质量提升了2.6倍，并在弗莱堡森林上达到了65.9的mIoU，优于PiCIE（45.3 IoU）和STEGO（57.6 IoU）等无监督基线。此外，ProGuT在类别无关的树干实例基准上优于现有的无监督方法。

英文摘要

Panoptic segmentation in forest environments is bottlenecked not by semantic quality but by instance separation; existing unsupervised panoptic approaches produce usable stuff maps but near-zero thing quality. Depth or flow-based instance discovery methods needs sensors that are not always available. We present ProGuT (Prototype Guided Training), which produces panoptic pseudo-labels without per-image training masks, needing only unlabeled images and one-time cluster-to-class mapping. ProGuT clusters CLIP patch features, then recovers trunk instances through multiscale geometric prior that falsifies non-trunk structures via structure-tensor. This is cheap compared to depth, flow or class-supervision methods to create pseudo labels. These are then used for downstream tasks which we evaluate against other unsupervised baselines. ProGuT achieves a Panoptic Quality (PQ) of 65.2 on Our-forest dataset (2.6x improvement over the initial pseudo-label quality) and reaches 65.9 mIoU on Freiburg Forest, outperforming unsupervised baselines like PiCIE (45.3 IoU) and STEGO(57.6IoU). Additionally, ProGuT outperforms existing unsupervised methods for class-agnostic trunk instance benchmark.

发表机构

• Robotics Research Lab, RPTU Kaiserslautern-Landau（机器人研究实验室，莱茵兰-普法尔茨凯泽斯劳滕大学）

机构由 AI 辅助整理，请以论文原文为准。