paper-with-me

Panoptic Segmentation 벤치마크

Panoptic Segmentation on COCO test-dev

114개 결과 · ⬇ CSV · JSON

PQ

27.2 35.27 43.35 51.42 59.5 2018-09 2026-09 JSIS-Net — 27.2 (2018-09-06) JSIS-Net — 27.2 (2018-09-06) JSIS-Net — 27.2 (2018-09-06) TASCNet — 40.7 (2018-12-04) TASCNet — 40.7 (2018-12-04) TASCNet — 40.7 (2018-12-04) AUNet (ResNext-152-FPN) — 46.5 (2018-12-10) AUNet (ResNet-152-FPN) — 45.5 (2018-12-10) AUNet (ResNet-101-FPN) — 45.2 (2018-12-10) AUNet (ResNext-152-FPN) — 46.5 (2018-12-10) AUNet (ResNet-152-FPN) — 45.5 (2018-12-10) AUNet (ResNet-101-FPN) — 45.2 (2018-12-10) AUNet (ResNext-152-FPN) — 46.5 (2018-12-10) AUNet (ResNet-152-FPN) — 45.5 (2018-12-10) AUNet (ResNet-101-FPN) — 45.2 (2018-12-10) Panoptic FPN — 40.9 (2019-01-08) Panoptic FPN — 40.9 (2019-01-08) Panoptic FPN — 40.9 (2019-01-08) UPSNet (ResNet-101-FPN) — 46.6 (2019-01-12) UPSNet (ResNet-101-FPN) — 46.6 (2019-01-12) UPSNet (ResNet-101-FPN) — 46.6 (2019-01-12) OCFusion (ResNeXt-101-FPN) — 46.6 (2019-06-13) OCFusion (ResNeXt-101-FPN) — 46.6 (2019-06-13) OCFusion (ResNeXt-101-FPN) — 46.6 (2019-06-13) GES Net — 33.7 (2019-08-24) GES Net — 33.7 (2019-08-24) GES Net — 33.7 (2019-08-24) AdaptIS (ResNeXt-101) — 42.8 (2019-09-17) AdaptIS (ResNeXt-101) — 42.8 (2019-09-17) AdaptIS (ResNeXt-101) — 42.8 (2019-09-17) SpatialFlow(ResNet-101-FPN) — 48.5 (2019-10-19) SpatialFlow(ResNet-101-FPN) — 48.5 (2019-10-19) SpatialFlow(ResNet-101-FPN) — 48.5 (2019-10-19) SOGNet (ResNet-101-FPN) — 47.8 (2019-11-18) SOGNet (ResNet-101-FPN) — 47.8 (2019-11-18) SOGNet (ResNet-101-FPN) — 47.8 (2019-11-18) Panoptic-DeepLab (Xception-71) — 41.4 (2019-11-22) Panoptic-DeepLab (Xception-71) — 41.4 (2019-11-22) Panoptic-DeepLab (Xception-71) — 41.4 (2019-11-22) Axial-DeepLab-L (multi-scale) — 44.2 (2020-03-17) Axial-DeepLab-L — 43.6 (2020-03-17) Axial-DeepLab-L (multi-scale) — 44.2 (2020-03-17) Axial-DeepLab-L — 43.6 (2020-03-17) Axial-DeepLab-L (multi-scale) — 44.2 (2020-03-17) Axial-DeepLab-L — 43.6 (2020-03-17) EPSNet (ResNet-101-FPN) — 38.9 (2020-03-23) EPSNet (ResNet-101-FPN) — 38.9 (2020-03-23) EPSNet (ResNet-101-FPN) — 38.9 (2020-03-23) PCV (ResNet-50) — 37.7 (2020-04-04) PCV (ResNet-50) — 37.7 (2020-04-04) PCV (ResNet-50) — 37.7 (2020-04-04) DetectoRS (ResNeXt-101-64x4d, multi-scale) — 50.0 (2020-06-03) DetectoRS (ResNeXt-101-64x4d, multi-scale) — 50.0 (2020-06-03) DetectoRS (ResNeXt-101-64x4d, multi-scale) — 50.0 (2020-06-03) Panoptic-DeepLab (SWideRNet-[1, 1, 4], multi-scale) — 46.5 (2020-11-23) Panoptic-DeepLab (SWideRNet-[1, 1, 4], multi-scale) — 46.5 (2020-11-23) Panoptic-DeepLab (SWideRNet-[1, 1, 4], multi-scale) — 46.5 (2020-11-23) Panoptic FCN* (Swin-L) — 52.7 (2020-12-01) MaX-DeepLab-L (single-scale) — 51.3 (2020-12-01) Panoptic FCN*++ (DCN-101-FPN) — 47.5 (2020-12-01) Panoptic FCN* (Swin-L) — 52.7 (2020-12-01) MaX-DeepLab-L (single-scale) — 51.3 (2020-12-01) Panoptic FCN*++ (DCN-101-FPN) — 47.5 (2020-12-01) Panoptic FCN* (Swin-L) — 52.7 (2020-12-01) MaX-DeepLab-L (single-scale) — 51.3 (2020-12-01) Panoptic FCN*++ (DCN-101-FPN) — 47.5 (2020-12-01) Ada-Segment (ResNet-101-DCN) — 48.5 (2020-12-07) Ada-Segment (ResNet-101-DCN) — 48.5 (2020-12-07) Ada-Segment (ResNet-101-DCN) — 48.5 (2020-12-07) REFINE (ResNeXt-101-DCN) — 51.5 (2020-12-15) REFINE (ResNet-101-DCN) — 49.6 (2020-12-15) REFINE (ResNeXt-101-DCN) — 51.5 (2020-12-15) REFINE (ResNet-101-DCN) — 49.6 (2020-12-15) REFINE (ResNeXt-101-DCN) — 51.5 (2020-12-15) REFINE (ResNet-101-DCN) — 49.6 (2020-12-15) COPS (ResNet-50) — 38.5 (2021-06-06) COPS (ResNet-50) — 38.5 (2021-06-06) COPS (ResNet-50) — 38.5 (2021-06-06) K-Net (Swin-L) — 55.2 (2021-06-28) K-Net (R101-FPN-DCN) — 48.3 (2021-06-28) K-Net (Swin-L) — 55.2 (2021-06-28) K-Net (R101-FPN-DCN) — 48.3 (2021-06-28) K-Net (Swin-L) — 55.2 (2021-06-28) K-Net (R101-FPN-DCN) — 48.3 (2021-06-28) MaskFormer (Swin-L) — 53.3 (2021-07-13) MaskFormer (Swin-L) — 53.3 (2021-07-13) MaskFormer (Swin-L) — 53.3 (2021-07-13) Panoptic SegFormer (Swin-L) — 56.2 (2021-09-08) Panoptic SegFormer (PVTv2-B5) — 55.8 (2021-09-08) Panoptic SegFormer (ResNet-101) — 50.9 (2021-09-08) Panoptic SegFormer (ResNet-50) — 50.2 (2021-09-08) Panoptic SegFormer (Swin-L) — 56.2 (2021-09-08) Panoptic SegFormer (PVTv2-B5) — 55.8 (2021-09-08) Panoptic SegFormer (ResNet-101) — 50.9 (2021-09-08) Panoptic SegFormer (ResNet-50) — 50.2 (2021-09-08) Panoptic SegFormer (Swin-L) — 56.2 (2021-09-08) Panoptic SegFormer (PVTv2-B5) — 55.8 (2021-09-08) Panoptic SegFormer (ResNet-101) — 50.9 (2021-09-08) Panoptic SegFormer (ResNet-50) — 50.2 (2021-09-08) Mask2Former (Swin-L) — 58.3 (2021-12-02) Mask2Former (Swin-L) — 58.3 (2021-12-02) Mask2Former (Swin-L) — 58.3 (2021-12-02) Mask DINO (single scale) — 59.5 (2022-06-06) Mask DINO (single scale) — 59.5 (2022-06-06) Mask DINO (single scale) — 59.5 (2022-06-06) CMT-DeepLab (single-scale) — 55.7 (2022-06-17) CMT-DeepLab (single-scale) — 55.7 (2022-06-17) CMT-DeepLab (single-scale) — 55.7 (2022-06-17) kMaX-DeepLab (single-scale) — 58.5 (2022-07-08) kMaX-DeepLab (single-scale) — 58.5 (2022-07-08) kMaX-DeepLab (single-scale) — 58.5 (2022-07-08) MaskConver (ResNet50, single-scale) — 53.6 (2023-12-11) MaskConver (ResNet50, single-scale) — 53.6 (2023-12-11) MaskConver (ResNet50, single-scale) — 53.6 (2023-12-11) JSIS-Net — 27.2 (2018-09-06) TASCNet — 40.7 (2018-12-04) AUNet (ResNext-152-FPN) — 46.5 (2018-12-10) UPSNet (ResNet-101-FPN) — 46.6 (2019-01-12) SpatialFlow(ResNet-101-FPN) — 48.5 (2019-10-19) DetectoRS (ResNeXt-101-64x4d, multi-scale) — 50.0 (2020-06-03) Panoptic FCN* (Swin-L) — 52.7 (2020-12-01) K-Net (Swin-L) — 55.2 (2021-06-28) Panoptic SegFormer (Swin-L) — 56.2 (2021-09-08) Mask2Former (Swin-L) — 58.3 (2021-12-02) Mask DINO (single scale) — 59.5 (2022-06-06)
RankModel PQPQstPQth PaperCodeYear
1 Mask DINO (single scale) 59.5-- Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation PaddlePaddle/PaddleDetection · IDEACVR/DINO · idea-research/maskdino · +7 2022
2 kMaX-DeepLab (single-scale) 58.549.064.8 kMaX-DeepLab: k-means Mask Transformer google-research/deeplab2 · bytedance/kmax-deeplab · cy-xu/spatially_aware_ai 2022
3 Mask2Former (Swin-L) 58.348.165.1 Masked-attention Mask Transformer for Universal Image Segmentation huggingface/transformers · open-mmlab/mmdetection · facebookresearch/Mask2Former · +4 2021
4 Panoptic SegFormer (Swin-L) 56.247.062.3 Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers zhiqi-li/Panoptic-SegFormer · claud1234/clft · claud1234/fcn_transformer_object_segmentation 2021
5 Panoptic SegFormer (PVTv2-B5) 55.846.561.9 Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers zhiqi-li/Panoptic-SegFormer · claud1234/clft · claud1234/fcn_transformer_object_segmentation 2021
6 CMT-DeepLab (single-scale) 55.746.861.6 CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation bytedance/kmax-deeplab · 2024-MindSpore-1/Code7 2022
7 K-Net (Swin-L) 55.246.261.2 K-Net: Towards Unified Image Segmentation zwwwayne/k-net 2021
8 MaskConver (ResNet50, single-scale) 53.658.945.6 MaskConver: Revisiting Pure Convolution Model for Panoptic Segmentation tensorflow/models 2023
9 MaskFormer (Swin-L) 53.344.559.1 Per-Pixel Classification is Not All You Need for Semantic Segmentation huggingface/transformers · open-mmlab/mmdetection · facebookresearch/MaskFormer 2021
10 Panoptic FCN* (Swin-L) 52.7 59.4 Fully Convolutional Networks for Panoptic Segmentation dvlab-research/panopticfcn · yanwei-li/PanopticFCN · Jia-Research-Lab/PanopticFCN · +3 2020
11 REFINE (ResNeXt-101-DCN) 51.539.259.6 REFINE: Prediction Fusion Network for Panoptic Segmentation 2020
12 MaX-DeepLab-L (single-scale) 51.342.457.2 MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers google-research/deeplab2 · conradry/max-deeplab · bytedance/kmax-deeplab 2020
13 Panoptic SegFormer (ResNet-101) 50.943.056.2 Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers zhiqi-li/Panoptic-SegFormer · claud1234/clft · claud1234/fcn_transformer_object_segmentation 2021
14 Panoptic SegFormer (ResNet-50) 50.242.455.3 Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers zhiqi-li/Panoptic-SegFormer · claud1234/clft · claud1234/fcn_transformer_object_segmentation 2021
15 DetectoRS (ResNeXt-101-64x4d, multi-scale) 5037.258.5 DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution open-mmlab/mmdetection · joe-siyuan-qiao/DetectoRS · FenHua/Robust_Logo_Detection · +3 2020
16 REFINE (ResNet-101-DCN) 49.637.757.5 REFINE: Prediction Fusion Network for Panoptic Segmentation 2020
17 SpatialFlow(ResNet-101-FPN) 48.537.955.5 SpatialFlow: Bridging All Tasks for Panoptic Segmentation chensnathan/SpatialFlow 2019
17 Ada-Segment (ResNet-101-DCN) 48.537.655.7 Ada-Segment: Automated Multi-loss Adaptation for Panoptic Segmentation 2020
19 K-Net (R101-FPN-DCN) 48.339.754 K-Net: Towards Unified Image Segmentation zwwwayne/k-net 2021
20 SOGNet (ResNet-101-FPN) 47.8 SOGNet: Scene Overlap Graph Network for Panoptic Segmentation LaoYang1994/SOGNet 2019
1–20 / 114 다음 → 페이지당 10 20 50 100