# Technical Report for the ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Pretraining-Diverse Ensemble of Foundation Vision Encoders for Robust Outdoor Scene Understanding

Source: [Technical Report for the ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge: Pretraining-Diverse Ensemble of Foundation Vision Encoders for Robust Outdoor Scene Understanding](<https://www.sota2.com/research/paper/6a3a13b6206542251d7c7886>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-24T05:00:31.418Z

Authors: Boyan Wang, Yongxi Huang, Wenjing Li, Tianrui Hui, Shaofei Huang, Nan Pu, Zhun Zhong

Published: 2026-06-22T09:57:55.000Z

## Abstract

This report presents our solution for the ICRA 2026 GOOSE 2D Fine-Grained Semantic Segmentation Challenge, which requires parsing unstructured outdoor scenes from four camera platforms into 56 fine-grained categories. Our approach pairs foundation vision encoders (including DINOv3, SigLIP2, and InternImage) with a Mask2Former decoder, and trains them with a strong recipe including long training schedules, exponential moving average, a larger crop size, and multi-scale plus flip test-time augmentation. The three encoders, chosen for their complementary pretraining objectives, are combined into a pretraining-diverse ensemble through per-class validation-IoU weighting. Evaluated on the official GOOSE test set, our submission achieves 75.40% composite mIoU and wins the second place of the challenge. Our study further shows that the encoder's pretraining recipe, rather than its parameter count or the decoder design, is the dominant factor for accuracy on this benchmark.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Semantic Segmentation on GOOSE (test)](<https://www.sota2.com/research/sota/semantic-segmentation-on-goose-test.md>): Semantic segmentation; GOOSE (test); mIoU (component): 75.4.

- [Semantic Segmentation on GOOSE (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-goose-val.md>): Semantic segmentation; GOOSE (val); Composite mIoU: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.23113>)

[PDF](<https://arxiv.org/pdf/2606.23113.pdf>)