HyperAIHyperAI

Command Palette

Search for a command to run...

Console

Panoptic Segmentation On Coco Test Dev

Metrics

PQ
PQst
PQth

Results

Performance results of various models on this benchmark

Paper Title
Mask DINO (single scale)59.5--Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation
kMaX-DeepLab (single-scale)58.549.064.8kMaX-DeepLab: k-means Mask Transformer
Mask2Former (Swin-L)58.348.165.1Masked-attention Mask Transformer for Universal Image Segmentation
Panoptic SegFormer (Swin-L)56.247.062.3Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
Panoptic SegFormer (PVTv2-B5)55.846.561.9Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
CMT-DeepLab (single-scale)55.746.861.6CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation
K-Net (Swin-L)55.246.261.2K-Net: Towards Unified Image Segmentation
MaskConver (ResNet50, single-scale)53.658.945.6MaskConver: Revisiting Pure Convolution Model for Panoptic Segmentation
MaskFormer (Swin-L)53.344.559.1Per-Pixel Classification is Not All You Need for Semantic Segmentation
Panoptic FCN* (Swin-L)52.7- 59.4Fully Convolutional Networks for Panoptic Segmentation
REFINE (ResNeXt-101-DCN)51.539.259.6REFINE: Prediction Fusion Network for Panoptic Segmentation
MaX-DeepLab-L (single-scale)51.342.457.2MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers
Panoptic SegFormer (ResNet-101)50.943.056.2Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
Panoptic SegFormer (ResNet-50)50.242.455.3Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers
DetectoRS (ResNeXt-101-64x4d, multi-scale)5037.258.5DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
REFINE (ResNet-101-DCN)49.637.757.5REFINE: Prediction Fusion Network for Panoptic Segmentation
Ada-Segment (ResNet-101-DCN)48.537.655.7Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation
SpatialFlow(ResNet-101-FPN)48.537.955.5SpatialFlow: Bridging All Tasks for Panoptic Segmentation
K-Net (R101-FPN-DCN)48.339.754K-Net: Towards Unified Image Segmentation
SOGNet (ResNet-101-FPN)47.8--SOGNet: Scene Overlap Graph Network for Panoptic Segmentation
0 of 38 row(s) selected.