Semantic labeling datasets results

Semantic labeling datasets results

What is the label of each pixel in the image ?

Discover the current state of the art in semantic labeling.

MSRC-21
who is the best in MSRC-21 ?

MSRC-21 12 results collected

Units: accuracy % per-class / (and) per-pixel

One of the oldest and classic dataset for semantic labelling. 21 different categories of surfaces are considered. Despite the innacuracies in the annotations and how unbalanced the classes are, this dataset still is commonly used as reference point. Note that here we consider the original annotations (where most results are published), not the cleaned-up version.

The results are reported per-class and per-pixel (this is sometimes called “average” and “global” result, respectively).

Original dataset website

Result Method Venue Details
80.9% / 86.8% Large-Scale Semantic Co-Labeling of Image Sets WACV 2014
80% / 83% Harmony Potentials - Fusing Local and Global Scale for Semantic Image Segmentation IJCV 2012 Details
79% / 86% Describing the Scene as a Whole: Joint Object Detection, Scene Classification and Semantic Segmentation CVPR 2012
78.2% / 85.0% Morphological Proximity Priors: Spatial Relationships for Semantic Segmentation TUM-I1222 2013
78% / 86% Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials NIPS 2011 Details
77% / 87% Graph Cut based Inference with Co-occurrence Statistics ECCV 2010
77% / 85% Are Spatial and Global Constraints Really Necessary for Segmentation? ICCV 2011 Details
76 % / 82 % Structured Image Segmentation using Kernelized Features ECCV 2012 Details
72.8% / 79.0% PatchMatchGraph: Building a Graph of Dense Patch Correspondences for Label Transfer ECCV 2012 Details
69% / 78% Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation PAMI 2010
67% / 72% Semantic Texton Forests for Image Categorization and Segmentation CVPR 2008
57% / 72% TextonBoost for Image Understanding IJCV 2009 Details
Something is off, something is missing ? Feel free to fill in the form.