Skip to main content
Frontiers in Plant Science logoLink to Frontiers in Plant Science
. 2026 Sep 21;17:1998555. doi: 10.3389/fpls.2026.1998555

Correction: SPINet: structure-enhanced point cloud instance segmentation network for multi-plant rice phenotyping

Wentao Song 1,2,†, Fang Qu 1,2,†, Yuwei Hao 1,2, He Huang 1,2,3, Caiguo Tang 1,2,3, Youqiang Sun 1,*, Junqing Zhang 1,*
PMCID: PMC13637161  PMID: 42836053

There was a mistake in Figure 10 as published. The image displayed for Figure 10 was inadvertently duplicated from Figure 5. The corrected Figure 10 appears below.

Figure 10.

Figure 10 presents a qualitative cross-dataset comparison of OneFormer3D and SPINet on 12 RicePanicle point-cloud samples. The samples are arranged in three groups of four columns, with OneFormer3D results shown in the upper row and the corresponding SPINet results shown directly below. Different colors denote individual predicted instances. In many samples, OneFormer3D produces large merged regions dominated by a single color, whereas SPINet produces more finely separated, multicolored grain-level instances across the rice panicles.

Qualitative cross-dataset comparison between OneFormer3D and SPINet on RicePanicle. Colors distinguish predicted instances within each result.

The original version of this article has been updated.

Footnotes

Edited and reviewed by: Frontiers Editorial Office, Frontiers Media SA, Switzerland

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.


Articles from Frontiers in Plant Science are provided here courtesy of Frontiers Media SA

RESOURCES