ImageNet-S

ImageNet Semantic Segmentation

Dataset Information
Modalities
Images
Introduced
2021
License
Unknown
Homepage
Official Website
Contents

Overview

Powered by the ImageNet dataset, unsupervised learning on large-scale data has made significant advances for classification tasks. There are two major challenges to allowing such an attractive learning modality for segmentation tasks: i) a large-scale benchmark for assessing algorithms is missing; ii) unsupervised shape representation learning is difficult. We propose a new problem of large-scale unsupervised semantic segmentation (LUSS) with a newly created benchmark dataset to track the research progress. Based on the ImageNet dataset, we propose the ImageNet-S dataset with 1.2 million training images and 50k high-quality semantic segmentation annotations for evaluation. Our benchmark has a high data diversity and a clear task objective. We also present a simple yet effective baseline method that works surprisingly well for LUSS. In addition, we benchmark related un/weakly/fully supervised methods accordingly, identifying the challenges and possible directions of LUSS.

Variants: ImageNet-S, ImageNet-S-300, ImageNet-S-50

Associated Benchmarks

This dataset is used in 4 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
Prompt Engineering MMRL MMRL: Multi-Modal Representation Learning for … 2025-03-11
Prompt Engineering HPT++ HPT++: Hierarchically Prompting Vision-Language Models … 2024-08-27
Prompt Engineering HPT Learning Hierarchical Prompt with Structured … 2023-12-11
Prompt Engineering PromptSRC Self-regulating Prompts: Foundational Model Adaptation … 2023-07-13
Prompt Engineering CoPrompt Consistency-guided Prompt Learning for Vision-Language … 2023-06-01
Prompt Engineering POMP Prompt Pre-Training with Twenty-Thousand Classes … 2023-04-10
Semantic Segmentation TEC (ViT-B/16, 224x224, SSL, mmseg) Towards Sustainable Self-supervised Learning 2022-10-20
Semantic Segmentation TEC (ViT-B/16, 224x224, SSL+FT) Towards Sustainable Self-supervised Learning 2022-10-20
Semantic Segmentation TEC (ViT-B/16, 224x224, SSL) Towards Sustainable Self-supervised Learning 2022-10-20
Semantic Segmentation TEC (ViT-B/16, 224x224, SSL+FT, mmseg) Towards Sustainable Self-supervised Learning 2022-10-20
Prompt Engineering MaPLe MaPLe: Multi-modal Prompt Learning 2022-10-06
Zero-Shot Transfer Image Classification PaLI PaLI: A Jointly-Scaled Multilingual Language-Image … 2022-09-14
Semantic Segmentation RF-ConvNext-Tiny (rfsingle, P4, 224x224, SUP) RF-Next: Efficient Receptive Field Search … 2022-06-14
Semantic Segmentation RF-ConvNext-Tiny (rfmerge, P4, 224x224, SUP) RF-Next: Efficient Receptive Field Search … 2022-06-14
Semantic Segmentation RF-ConvNext-Tiny (rfmultiple, P4, 224x224, SUP) RF-Next: Efficient Receptive Field Search … 2022-06-14
Semantic Segmentation SERE (ViT-B/16, 100ep, 224x224, SSL+FT) SERE: Exploring Feature Self-relation for … 2022-06-10
Semantic Segmentation SERE (ViT-S/16, 100ep, 224x224, SSL) SERE: Exploring Feature Self-relation for … 2022-06-10
Semantic Segmentation SERE (ViT-S/16, 100ep, 224x224, SSL+FT) SERE: Exploring Feature Self-relation for … 2022-06-10
Semantic Segmentation SERE (ViT-S/16, 100ep, 224x224, SSL+FT, mmseg) SERE: Exploring Feature Self-relation for … 2022-06-10
Semantic Segmentation SERE (ViT-B/16, 100ep, 224x224, SSL) SERE: Exploring Feature Self-relation for … 2022-06-10

Research Papers

Recent papers with results on this dataset:

External Links: