SBD

Semantic Boundaries Dataset

Dataset Information
Modalities
Images
Introduced
2011
License
Unknown
Homepage
Official Website
Contents

Overview

The Semantic Boundaries Dataset (SBD) is a dataset for predicting pixels on the boundary of the object (as opposed to the inside of the object with semantic segmentation). The dataset consists of 11318 images from the trainval set of the PASCAL VOC2011 challenge, divided into 8498 training and 2820 test images. This dataset has object instance boundaries with accurate figure/ground masks that are also labeled with one of 20 Pascal VOC classes.

Source: Weakly Supervised Object Boundaries
Image Source: http://home.bharathh.info/pubs/codes/SBD/download.html

Variants: SBD, Sbd val

Associated Benchmarks

This dataset is used in 2 benchmarks:

Recent Benchmark Submissions

Task Model Paper Date
Interactive Segmentation ViT-B+MST+CL MST: Adaptive Multi-Scale Tokens Guided … 2024-01-09
Interactive Segmentation SimpleClick CFR-1 (ViT-H, SBD) CFR-ICL: Cascade-Forward Refinement with Iterative … 2023-03-09
Interactive Segmentation SimpleClick SimpleClick: Interactive Image Segmentation with … 2022-10-20
Interactive Segmentation FocalClick FocalClick: Towards Practical Interactive Image … 2022-04-06
Interactive Segmentation IA-FP-Net(HRNet,SBD) Cascaded Sparse Feature Propagation Network … 2022-03-10
Interactive Segmentation UCP-Net UCP-Net: Unstructured Contour Points for … 2021-09-15
Interactive Segmentation RITM (HRNet18, SBD) Reviving Iterative Training with Mask … 2021-02-12
Interactive Segmentation f-BRS-B (ResNet-101) f-BRS: Rethinking Backpropagating Refinement for … 2020-01-28
Edge Detection CASENet CASENet: Deep Category-Aware Semantic Edge … 2017-05-27
Interactive Segmentation DOS with GC Deep Interactive Object Selection 2016-03-13
Interactive Segmentation DOS w/o GC Deep Interactive Object Selection 2016-03-13
Edge Detection WSOB Weakly Supervised Object Boundaries 2015-11-24

Research Papers

Recent papers with results on this dataset:

External Links: