AI Paper Atlas

Research map / Benchmarks / COCO val2017

COCO val2017 leaderboard

82 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on COCO val2017. Best AP: 79.3 (PCT, CVPR 2023).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

COCO val2017 · AP

66 papers, each with its best reported number for its own method; higher is better.

56 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

COCO val2017 · mask AP

17 papers, each with its best reported number for its own method; higher is better.

#mask AP ↑MethodPaperTopic
144.2EXITS
Swin-Small + Mask2Former
Extreme Point Supervised Instance Segmentation
CVPR 2024 · Table 1
Weakly supervised detection and segmentation
243.3Mask Auto-Labeler (ViT-MAE-Base labeler, Swin-Small Mask2Former)
ViT-MAE-Base
Vision Transformers Are Good Mask Auto-Labelers
CVPR 2023 · Table 1
Weakly supervised detection and segmentation
342.1HTC + RFP + SAC
ResNet-50
DetectoRS: Detecting Objects With Recursive Feature Pyramid and Switchable Atrous Convolution
CVPR 2021 · Table: detection results with
CNN features
441.2DCT-Mask R-CNN
ResNeXt-101
DCT-Mask: Discrete Cosine Transform Mask Representation for Instance Segmentation
CVPR 2021 · Backbone comparison table
CNN features
540.9PointRend, X101-FPN
X101-FPN
PointRend: Image Segmentation As Rendering
CVPR 2020 · Table 4
CNN features
638.4TransMix-PVT-S Mask R-CNN
PVT-S
TransMix: Attend To Mix for Vision Transformers
CVPR 2022 · Table 3
CNN features
737.8ReXNet (×2.2) + FPN
ResNet50
Rethinking Channel Dimensions for Efficient Model Design
CVPR 2021 · Table 11
CNN features
837.8PANet
ResNet-50
Path Aggregation Network for Instance Segmentation
CVPR 2018 · Ablation table
CNN features
937.1Ours with tag
ResNet-50-FPN
Multimodal Contrastive Training for Visual Representation Learning
CVPR 2021 · Table: Instance Segmentation o
Cross-modal learning
1037.1Mask R-CNN +1 NL
ResNet-101
Non-Local Neural Networks
CVPR 2018 · Table: COCO object detection a
CNN features
7 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.