AI Paper Atlas

Research map / Benchmarks / Potsdam

Potsdam leaderboard

12 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Potsdam. Best mIoU: 87.71 (UNet-ConvNeXt(S)-GMamba, ICLR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

Potsdam · mIoU

12 papers, each with its best reported number for its own method; higher is better.

#mIoU ↑MethodPaperTopic
187.71UNet-ConvNeXt(S)-GMamba
UNet-ConvNeXt(S)
Enabling True Global Perception in State Space Models for Visual Tasks
ICLR 2026 · Table 1
CNN features
287.2LSKNet-S
UNetFormer
LSKNet: A Foundation Lightweight Backbone for Remote Sensing
IJCV 2024 · Table 8
Human and embodied modeling
382.76SimCLR+ResNet50+UperNet
ResNet50
Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images (Student Abstract)
AAAI 2023 · Table 2
Self-supervised and contrastive
475.4PFNet
ResNet50
PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation
CVPR 2021 · Potsdam results table
CNN features
559.69SynRS3DSynRS3D: A Synthetic Dataset for Global 3D Semantic Understanding from Monocular Remote Sensing Imagery
NeurIPS 2024 · Source-only land cover mapping
3D generation
658.9Ours
DINOv2-ViT-B14-REG
Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure
NeurIPS 2024 · Table 1
Open-set and few-shot detection
754.1SkySense-OSkySense-O: Towards Open-World Remote Sensing Interpretation with Vision-Centric Visual-Language Modeling
CVPR 2025 · Table 2
CLIP and zero-shot
851.1TMPA
SegEarth-OV (CLIP ViT-B/16)
Test-Time Multi-Prompt Adaptation for Open-Vocabulary Remote Sensing Image Segmentation
CVPR 2026 · Table 1
Open-set and few-shot detection
948.5VPTA
CLIP ViT-B/16 + SegEarth-OV
Label-Free Text Prototype Adaptation for Open Vocabulary Segmentation
ECCV 2026 · Table 1
Domain adaptation and generalization
1047.1SegEarth-OV
CLIP ViT-B/16
SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images
CVPR 2025 · Table 1
Open-set and few-shot detection
2 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.