16k on COCO minival

Metric: APL (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide augmentations

Sort:

#	Model↕	APL▼	Augmentations	Paper	Date↕	Code
1	Focal-Stable-DINO (Focal-Huge, no TTA)	78.5	Yes	A Strong and Reproducible Object Detector with O...	2023-04-25	Code
2	EVA	78.5	Yes	EVA: Exploring the Limits of Masked Visual Repre...	2022-11-14	Code
3	UNINEXT-H	75.3	Yes	Universal Instance Perception as Object Discover...	2023-03-12	Code
4	DyHead (Swin-L, multi scale, self-training)	74.2	Yes	Dynamic Head: Unifying Object Detection Heads wi...	2021-06-15	Code
5	Focal-L (DyHead, multi-scale)	73.4	No	Focal Self-attention for Local-Global Interactio...	2021-07-01	Code
6	DyHead (Swin-L, multi scale)	73.2	No	Dynamic Head: Unifying Object Detection Heads wi...	2021-06-15	Code
7	SOLQ (Swin-L, single scale)	71.9	No	SOLQ: Segmenting Objects by Learning Queries	2021-06-04	Code
8	QueryInst (single scale)	71.5	No	Instances as Queries	2021-05-05	Code
9	Cascade RCNN-RS (SpineNet-143L, single scale)	70.6	No	Simple Training Strategies and Model Scaling for...	2021-06-30	Code
10	Cascade RCNN-RS (ResNet-200, single scale)	70.3	No	Simple Training Strategies and Model Scaling for...	2021-06-30	Code
11	YOLOR-D6 (1280, single-scale, 31 fps)	68.7	No	You Only Learn One Representation: Unified Netwo...	2021-05-10	Code
12	UniverseNet-20.08d (Res2Net-101, DCN, multi-scale)	68.1	No	USB: Universal-Scale Object Detection Benchmark	2021-03-25	Code
13	EfficientDet-D7x (single-scale)	67.9	No	EfficientDet: Scalable and Efficient Object Dete...	2019-11-20	Code
14	YOLOv4-P7 CSP-P7 (single-scale, 16 fps)	67.4	No	Scaled-YOLOv4: Scaling Cross Stage Partial Network	2020-11-16	Code
15	DyHead (ResNeXt-64x4d-101-DCN, multi scale)	66.3	No	Dynamic Head: Unifying Object Detection Heads wi...	2021-06-15	Code
16	ResNeSt-200 (multi-scale)	66.29	No	ResNeSt: Split-Attention Networks	2020-04-19	Code
17	ResNeSt-200-DCN (single-scale)	65.83	No	ResNeSt: Split-Attention Networks	2020-04-19	Code
18	DINO-5scale (24 epoch)	65.8	No	DINO: DETR with Improved DeNoising Anchor Boxes ...	2022-03-07	Code
19	UniverseNet-20.08d (Res2Net-101, DCN, single-scale)	65.8	No	USB: Universal-Scale Object Detection Benchmark	2021-03-25	Code
20	DN-Deformable-DETR-R50++	65.4	No	DN-DETR: Accelerate DETR Training by Introducing...	2022-03-02	Code
21	DINO-5scale (36 epoch)	65.3	No	DINO: DETR with Improved DeNoising Anchor Boxes ...	2022-03-07	Code
22	YOLOR-P6 (1280, single-scale, 72 fps)	65.2	No	You Only Learn One Representation: Unified Netwo...	2021-05-10	Code
23	REGO-Deformable DETR-X101	65	No	Recurrent Glimpse-based Decoder for Detection wi...	2021-12-09	Code
24	DAB-DETR-DC5-R101	64.1	No	DAB-DETR: Dynamic Anchor Boxes are Better Querie...	2022-01-28	Code
25	ResNeSt-200 (single-scale)	63.9	No	ResNeSt: Split-Attention Networks	2020-04-19	Code
26	Conditional DETR-R101	63.6	No	Conditional DETR for Fast Training Convergence	2021-08-13	Code
27	Conditional DETR-DC5-R101	63.3	No	Conditional DETR for Fast Training Convergence	2021-08-13	Code
28	DAB-DETR-R101	62.9	No	DAB-DETR: Dynamic Anchor Boxes are Better Querie...	2022-01-28	Code
29	UniverseNet-20.08 (Res2Net-50, DCN, single-scale)	62.7	No	USB: Universal-Scale Object Detection Benchmark	2021-03-25	Code
30	DETR-DC5 (ResNet-101)	62.3	No	End-to-End Object Detection with Transformers	2020-05-26	Code
31	HTC (HRNetV2p-W48)	62.2	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
32	Conditional DETR-DC5-R50	62.2	No	Conditional DETR for Fast Training Convergence	2021-08-13	Code
33	Res2Net101+HTC	62.1	No	Res2Net: A New Multi-scale Backbone Architecture	2019-04-02	Code
34	Sparse R-CNN (ResNet-101, learnable proposals, random crop aug, FPN)	61.6	No	Sparse R-CNN: End-to-End Object Detection with L...	2020-11-25	Code
35	Anchor DETR-DC5-R101	61.6	No	Anchor DETR: Query Design for Transformer-Based ...	2021-09-15	Code
36	Conditional DETR-R50	61.5	No	Conditional DETR for Fast Training Convergence	2021-08-13	Code
37	MAE-Det(MAE-Det-L+GFLV2)	61.1	No	MAE-DET: Revisiting Maximum Entropy Principle in...	2021-11-26	Code
38	Anchor DETR-DC5-R50	60.6	No	Anchor DETR: Query Design for Transformer-Based ...	2021-09-15	Code
39	Pix2seq (R101-DC5)	60.4	No	Pix2seq: A Language Modeling Framework for Objec...	2021-09-22	Code
40	Mask R-CNN (HRNetV2p-W48, cascade)	60.1	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
41	HoughNet (HG-104, MS)	59.7	No	HoughNet: Integrating near and long-range eviden...	2020-07-05	Code
42	Sparse R-CNN (ResNet-101, FPN)	59.7	No	Sparse R-CNN: End-to-End Object Detection with L...	2020-11-25	Code
43	R3-CNN (ResNet-50-FPN, DCN)	59.6	No	Recursively Refined R-CNN: Instance Segmentation...	2021-04-03	Code
44	HTC (HRNetV2p-W32)	59.5	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
45	Sparse R-CNN (ResNet-50, learnable proposals, random crop aug, FPN)	59.5	No	Sparse R-CNN: End-to-End Object Detection with L...	2020-11-25	Code
46	PVT-Large (RetinaNet 3x,MS)	59.5	No	Pyramid Vision Transformer: A Versatile Backbone...	2021-02-24	Code
47	ExtremeNet (Hourglass-104, multi-scale)	59.4	No	Bottom-up Object Detection by Grouping Extreme a...	2019-01-23	Code
48	R3-CNN (ResNet-50-FPN, GC-Net)	58.9	No	Recursively Refined R-CNN: Instance Segmentation...	2021-04-03	Code
49	CenterMask+VoVNetV2-99 (single-scale)	58.8	No	CenterMask : Real-Time Anchor-Free Instance Segm...	2019-11-15	Code
50	Faster R-CNN (ResNet-101, DCNv2)	58.7	No	Deformable ConvNets v2: More Deformable, Better ...	2018-11-27	Code
51	Pix2seq (R50-DC5 )	58.6	No	Pix2seq: A Language Modeling Framework for Objec...	2021-09-22	Code
52	Cascade R-CNN (HRNetV2p-W48)	58.5	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
53	PVT-Large (RetinaNet 1x)	58.4	No	Pyramid Vision Transformer: A Versatile Backbone...	2021-02-24	Code
54	CornerNet-Saccade (Hourglass-54)	58.4	No	CornerNet-Lite: Efficient Keypoint Based Object ...	2019-04-18	Code
55	RetinaNet (ViL-Base)	58.3	No	Multi-Scale Vision Longformer: A New Vision Tran...	2021-03-29	Code
56	RetinaNet (ViL-Base, multi-scale, 3x)	58.1	No	Multi-Scale Vision Longformer: A New Vision Tran...	2021-03-29	Code
57	Mask R-CNN (VoVNetV2-99, single-scale)	57.7	No	CenterMask : Real-Time Anchor-Free Instance Segm...	2019-11-15	Code
58	Sparse R-CNN (ResNet-50, FPN)	57.6	No	Sparse R-CNN: End-to-End Object Detection with L...	2020-11-25	Code
59	Cascade R-CNN (HRNetV2p-W32)	57.4	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
60	Cascade R-CNN (ResNet-101-FPN+, cascade)	57.4	No	Cascade R-CNN: Delving into High Quality Object ...	2017-12-03	Code
61	CenterMask+X101-32x8d (single-scale)	57.1	No	CenterMask : Real-Time Anchor-Free Instance Segm...	2019-11-15	Code
62	CornerNet-Saccade (Hourglass-104)	57.1	No	CornerNet-Lite: Efficient Keypoint Based Object ...	2019-04-18	Code
63	TridentNet (ResNet-101)	56.9	No	Scale-Aware Trident Networks for Object Detection	2019-01-07	Code
64	Mask R-CNN-FPN (ResNeXt-101, GN+WS)	56.39	No	Micro-Batch Training with Batch-Channel Normaliz...	2019-03-25	Code
65	ExtremeNet (Hourglass-104, single-scale)	56.1	No	Bottom-up Object Detection by Grouping Extreme a...	2019-01-23	Code
66	Faster RCNN-R101-FPN+	56	No	End-to-End Object Detection with Transformers	2020-05-26	Code
67	HoughNet (HG-104)	55.8	No	HoughNet: Integrating near and long-range eviden...	2020-07-05	Code
68	CenterNet511 (Hourglass-52)	55.8	No	CenterNet: Keypoint Triplets for Object Detection	2019-04-17	Code
69	R3-CNN (ResNet-50-FPN)	55.7	No	Recursively Refined R-CNN: Instance Segmentation...	2021-04-03	Code
70	Faster R-CNN (FPN, X-volution)	55	No	X-volution: On the unification of convolution an...	2021-06-04	-
71	Faster R-CNN (HRNetV2p-W48)	54.6	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
72	Grid R-CNN (ResNet-101-FPN)	54.1	No	Grid R-CNN	2018-11-29	Code
73	Cascade R-CNN (HRNetV2p-W18)	54.1	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
74	Cascade R-CNN (ResNet-50-FPN+)	54.1	No	Cascade R-CNN: Delving into High Quality Object ...	2017-12-03	Code
75	Faster R-CNN (HRNetV2p-W32)	53.3	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
76	FoveaBox (ResNet-101-FPN, 600x600)	52.7	No	FoveaBox: Beyond Anchor-based Object Detector	2019-04-08	Code
77	FPN+	52.6	No	Feature Pyramid Networks for Object Detection	2016-12-09	Code
78	GCnet (ResNet-50-FPN, GRoIE)	52.5	No	GCNet: Non-local Networks Meet Squeeze-Excitatio...	2019-04-25	Code
79	HTC (cascade)	52.3	No	Hybrid Task Cascade for Instance Segmentation	2019-01-22	Code
80	PPDet (ResNet-101-FPN)	52.3	No	Reducing Label Noise in Anchor-Free Object Detec...	2020-08-03	Code
81	CornerNet511 (Hourglass-104)	51.8	No	CornerNet: Detecting Objects as Paired Keypoints	2018-08-03	Code
82	FoveaBox (ResNet-101-FPN, 800x800)	51.7	No	FoveaBox: Beyond Anchor-based Object Detector	2019-04-08	Code
83	Grid R-CNN (ResNet-50-FPN)	51.5	No	Grid R-CNN	2018-11-29	Code
84	Faster R-CNN (Res2Net-50)	51.1	No	Res2Net: A New Multi-scale Backbone Architecture	2019-04-02	Code
85	Mask R-CNN (HRNetV2p-W18)	51	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
86	Libra R-CNN (ResNet-50 FPN)	50.5	No	Libra R-CNN: Towards Balanced Learning for Objec...	2019-04-04	Code
87	FoveaBox (ResNet-50-FPN, 600x600)	50.5	No	FoveaBox: Beyond Anchor-based Object Detector	2019-04-08	Code
88	FCOS (ResNet-50-FPN + improvements)	49.8	No	FCOS: Fully Convolutional One-Stage Object Detec...	2019-04-02	Code
89	Mask R-CNN (ResNet-50-FPN, GRoIE)	49.7	No	A novel Region of Interest Extraction Layer for ...	2020-04-28	Code
90	Faster R-CNN (HRNetV2p-W18)	49.6	No	Deep High-Resolution Representation Learning for...	2019-08-20	Code
91	M2Det (ResNet-1o1, 320x320)	49.3	No	M2Det: A Single-Shot Object Detector based on Mu...	2018-11-12	Code
92	M2Det (VGG-16, 320x320)	49.1	No	M2Det: A Single-Shot Object Detector based on Mu...	2018-11-12	Code
93	FSAF (ResNet-50)	48.2	No	Feature Selective Anchor-Free Module for Single-...	2019-03-02	Code
94	Faster R-CNN (ResNet-50-FPN, GRoIE)	47.8	No	A novel Region of Interest Extraction Layer for ...	2020-04-28	Code
95	GHM-C + GHM-R (RetinaNet-FPN-ResNet-50, M=30)	46.7	No	Gradient Harmonized Single-stage Detector	2018-11-13	Code

#1Focal-Stable-DINO (Focal-Huge, no TTA)
78.5
APL· Augmentations· 2023-04-25
A Strong and Reproducible Object Detector with Only Public Datasets Code
#2EVASOTA
78.5
APL· Augmentations· 2022-11-14
EVA: Exploring the Limits of Masked Visual Representation Learning at Scale Code
#3UNINEXT-H
75.3
APL· Augmentations· 2023-03-12
Universal Instance Perception as Object Discovery and Retrieval Code
#4DyHead (Swin-L, multi scale, self-training)SOTA
74.2
APL· Augmentations· 2021-06-15
Dynamic Head: Unifying Object Detection Heads with Attentions Code
#5Focal-L (DyHead, multi-scale)
73.4
APL· 2021-07-01
Focal Self-attention for Local-Global Interactions in Vision Transformers Code
#6DyHead (Swin-L, multi scale)
73.2
APL· 2021-06-15
Dynamic Head: Unifying Object Detection Heads with Attentions Code
#7SOLQ (Swin-L, single scale)SOTA
71.9
APL· 2021-06-04
SOLQ: Segmenting Objects by Learning Queries Code
#8QueryInst (single scale)SOTA
71.5
APL· 2021-05-05
Instances as Queries Code
#9Cascade RCNN-RS (SpineNet-143L, single scale)
70.6
APL· 2021-06-30
Simple Training Strategies and Model Scaling for Object Detection Code
#10Cascade RCNN-RS (ResNet-200, single scale)
70.3
APL· 2021-06-30
Simple Training Strategies and Model Scaling for Object Detection Code
#11YOLOR-D6 (1280, single-scale, 31 fps)
68.7
APL· 2021-05-10
You Only Learn One Representation: Unified Network for Multiple Tasks Code
#12UniverseNet-20.08d (Res2Net-101, DCN, multi-scale)SOTA
68.1
APL· 2021-03-25
USB: Universal-Scale Object Detection Benchmark Code
#13EfficientDet-D7x (single-scale)SOTA
67.9
APL· 2019-11-20
EfficientDet: Scalable and Efficient Object Detection Code
#14YOLOv4-P7 CSP-P7 (single-scale, 16 fps)
67.4
APL· 2020-11-16
Scaled-YOLOv4: Scaling Cross Stage Partial Network Code
#15DyHead (ResNeXt-64x4d-101-DCN, multi scale)
66.3
APL· 2021-06-15
Dynamic Head: Unifying Object Detection Heads with Attentions Code
#16ResNeSt-200 (multi-scale)
66.29
APL· 2020-04-19
ResNeSt: Split-Attention Networks Code
#17ResNeSt-200-DCN (single-scale)
65.83
APL· 2020-04-19
ResNeSt: Split-Attention Networks Code
#18DINO-5scale (24 epoch)
65.8
APL· 2022-03-07
DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection Code
#19UniverseNet-20.08d (Res2Net-101, DCN, single-scale)
65.8
APL· 2021-03-25
USB: Universal-Scale Object Detection Benchmark Code
#20DN-Deformable-DETR-R50++
65.4
APL· 2022-03-02
DN-DETR: Accelerate DETR Training by Introducing Query DeNoising Code
#21DINO-5scale (36 epoch)
65.3
APL· 2022-03-07
DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detection Code
#22YOLOR-P6 (1280, single-scale, 72 fps)
65.2
APL· 2021-05-10
You Only Learn One Representation: Unified Network for Multiple Tasks Code
#23REGO-Deformable DETR-X101
65
APL· 2021-12-09
Recurrent Glimpse-based Decoder for Detection with Transformer Code
#24DAB-DETR-DC5-R101
64.1
APL· 2022-01-28
DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR Code
#25ResNeSt-200 (single-scale)
63.9
APL· 2020-04-19
ResNeSt: Split-Attention Networks Code
#26Conditional DETR-R101
63.6
APL· 2021-08-13
Conditional DETR for Fast Training Convergence Code
#27Conditional DETR-DC5-R101
63.3
APL· 2021-08-13
Conditional DETR for Fast Training Convergence Code
#28DAB-DETR-R101
62.9
APL· 2022-01-28
DAB-DETR: Dynamic Anchor Boxes are Better Queries for DETR Code
#29UniverseNet-20.08 (Res2Net-50, DCN, single-scale)
62.7
APL· 2021-03-25
USB: Universal-Scale Object Detection Benchmark Code
#30DETR-DC5 (ResNet-101)
62.3
APL· 2020-05-26
End-to-End Object Detection with Transformers Code
#31HTC (HRNetV2p-W48)SOTA
62.2
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#32Conditional DETR-DC5-R50
62.2
APL· 2021-08-13
Conditional DETR for Fast Training Convergence Code
#33Res2Net101+HTCSOTA
62.1
APL· 2019-04-02
Res2Net: A New Multi-scale Backbone Architecture Code
#34Sparse R-CNN (ResNet-101, learnable proposals, random crop aug, FPN)
61.6
APL· 2020-11-25
Sparse R-CNN: End-to-End Object Detection with Learnable Proposals Code
#35Anchor DETR-DC5-R101
61.6
APL· 2021-09-15
Anchor DETR: Query Design for Transformer-Based Object Detection Code
#36Conditional DETR-R50
61.5
APL· 2021-08-13
Conditional DETR for Fast Training Convergence Code
#37MAE-Det(MAE-Det-L+GFLV2)
61.1
APL· 2021-11-26
MAE-DET: Revisiting Maximum Entropy Principle in Zero-Shot NAS for Efficient Object Detection Code
#38Anchor DETR-DC5-R50
60.6
APL· 2021-09-15
Anchor DETR: Query Design for Transformer-Based Object Detection Code
#39Pix2seq (R101-DC5)
60.4
APL· 2021-09-22
Pix2seq: A Language Modeling Framework for Object Detection Code
#40Mask R-CNN (HRNetV2p-W48, cascade)
60.1
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#41HoughNet (HG-104, MS)
59.7
APL· 2020-07-05
HoughNet: Integrating near and long-range evidence for bottom-up object detection Code
#42Sparse R-CNN (ResNet-101, FPN)
59.7
APL· 2020-11-25
Sparse R-CNN: End-to-End Object Detection with Learnable Proposals Code
#43R3-CNN (ResNet-50-FPN, DCN)
59.6
APL· 2021-04-03
Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing Code
#44HTC (HRNetV2p-W32)
59.5
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#45Sparse R-CNN (ResNet-50, learnable proposals, random crop aug, FPN)
59.5
APL· 2020-11-25
Sparse R-CNN: End-to-End Object Detection with Learnable Proposals Code
#46PVT-Large (RetinaNet 3x,MS)
59.5
APL· 2021-02-24
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions Code
#47ExtremeNet (Hourglass-104, multi-scale)SOTA
59.4
APL· 2019-01-23
Bottom-up Object Detection by Grouping Extreme and Center Points Code
#48R3-CNN (ResNet-50-FPN, GC-Net)
58.9
APL· 2021-04-03
Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing Code
#49CenterMask+VoVNetV2-99 (single-scale)
58.8
APL· 2019-11-15
CenterMask : Real-Time Anchor-Free Instance Segmentation Code
#50Faster R-CNN (ResNet-101, DCNv2)SOTA
58.7
APL· 2018-11-27
Deformable ConvNets v2: More Deformable, Better Results Code
#51Pix2seq (R50-DC5 )
58.6
APL· 2021-09-22
Pix2seq: A Language Modeling Framework for Object Detection Code
#52Cascade R-CNN (HRNetV2p-W48)
58.5
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#53PVT-Large (RetinaNet 1x)
58.4
APL· 2021-02-24
Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions Code
#54CornerNet-Saccade (Hourglass-54)
58.4
APL· 2019-04-18
CornerNet-Lite: Efficient Keypoint Based Object Detection Code
#55RetinaNet (ViL-Base)
58.3
APL· 2021-03-29
Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encoding Code
#56RetinaNet (ViL-Base, multi-scale, 3x)
58.1
APL· 2021-03-29
Multi-Scale Vision Longformer: A New Vision Transformer for High-Resolution Image Encoding Code
#57Mask R-CNN (VoVNetV2-99, single-scale)
57.7
APL· 2019-11-15
CenterMask : Real-Time Anchor-Free Instance Segmentation Code
#58Sparse R-CNN (ResNet-50, FPN)
57.6
APL· 2020-11-25
Sparse R-CNN: End-to-End Object Detection with Learnable Proposals Code
#59Cascade R-CNN (HRNetV2p-W32)
57.4
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#60Cascade R-CNN (ResNet-101-FPN+, cascade)SOTA
57.4
APL· 2017-12-03
Cascade R-CNN: Delving into High Quality Object Detection Code
#61CenterMask+X101-32x8d (single-scale)
57.1
APL· 2019-11-15
CenterMask : Real-Time Anchor-Free Instance Segmentation Code
#62CornerNet-Saccade (Hourglass-104)
57.1
APL· 2019-04-18
CornerNet-Lite: Efficient Keypoint Based Object Detection Code
#63TridentNet (ResNet-101)
56.9
APL· 2019-01-07
Scale-Aware Trident Networks for Object Detection Code
#64Mask R-CNN-FPN (ResNeXt-101, GN+WS)
56.39
APL· 2019-03-25
Micro-Batch Training with Batch-Channel Normalization and Weight Standardization Code
#65ExtremeNet (Hourglass-104, single-scale)
56.1
APL· 2019-01-23
Bottom-up Object Detection by Grouping Extreme and Center Points Code
#66Faster RCNN-R101-FPN+
56
APL· 2020-05-26
End-to-End Object Detection with Transformers Code
#67HoughNet (HG-104)
55.8
APL· 2020-07-05
HoughNet: Integrating near and long-range evidence for bottom-up object detection Code
#68CenterNet511 (Hourglass-52)
55.8
APL· 2019-04-17
CenterNet: Keypoint Triplets for Object Detection Code
#69R3-CNN (ResNet-50-FPN)
55.7
APL· 2021-04-03
Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing Code
#70Faster R-CNN (FPN, X-volution)
55
APL· 2021-06-04
X-volution: On the unification of convolution and self-attention
#71Faster R-CNN (HRNetV2p-W48)
54.6
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#72Grid R-CNN (ResNet-101-FPN)
54.1
APL· 2018-11-29
Grid R-CNN Code
#73Cascade R-CNN (HRNetV2p-W18)
54.1
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#74Cascade R-CNN (ResNet-50-FPN+)
54.1
APL· 2017-12-03
Cascade R-CNN: Delving into High Quality Object Detection Code
#75Faster R-CNN (HRNetV2p-W32)
53.3
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#76FoveaBox (ResNet-101-FPN, 600x600)
52.7
APL· 2019-04-08
FoveaBox: Beyond Anchor-based Object Detector Code
#77FPN+SOTA
52.6
APL· 2016-12-09
Feature Pyramid Networks for Object Detection Code
#78GCnet (ResNet-50-FPN, GRoIE)
52.5
APL· 2019-04-25
GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond Code
#79HTC (cascade)
52.3
APL· 2019-01-22
Hybrid Task Cascade for Instance Segmentation Code
#80PPDet (ResNet-101-FPN)
52.3
APL· 2020-08-03
Reducing Label Noise in Anchor-Free Object Detection Code
#81CornerNet511 (Hourglass-104)
51.8
APL· 2018-08-03
CornerNet: Detecting Objects as Paired Keypoints Code
#82FoveaBox (ResNet-101-FPN, 800x800)
51.7
APL· 2019-04-08
FoveaBox: Beyond Anchor-based Object Detector Code
#83Grid R-CNN (ResNet-50-FPN)
51.5
APL· 2018-11-29
Grid R-CNN Code
#84Faster R-CNN (Res2Net-50)
51.1
APL· 2019-04-02
Res2Net: A New Multi-scale Backbone Architecture Code
#85Mask R-CNN (HRNetV2p-W18)
51
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#86Libra R-CNN (ResNet-50 FPN)
50.5
APL· 2019-04-04
Libra R-CNN: Towards Balanced Learning for Object Detection Code
#87FoveaBox (ResNet-50-FPN, 600x600)
50.5
APL· 2019-04-08
FoveaBox: Beyond Anchor-based Object Detector Code
#88FCOS (ResNet-50-FPN + improvements)
49.8
APL· 2019-04-02
FCOS: Fully Convolutional One-Stage Object Detection Code
#89Mask R-CNN (ResNet-50-FPN, GRoIE)
49.7
APL· 2020-04-28
A novel Region of Interest Extraction Layer for Instance Segmentation Code
#90Faster R-CNN (HRNetV2p-W18)
49.6
APL· 2019-08-20
Deep High-Resolution Representation Learning for Visual Recognition Code
#91M2Det (ResNet-1o1, 320x320)
49.3
APL· 2018-11-12
M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network Code
#92M2Det (VGG-16, 320x320)
49.1
APL· 2018-11-12
M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network Code
#93FSAF (ResNet-50)
48.2
APL· 2019-03-02
Feature Selective Anchor-Free Module for Single-Shot Object Detection Code
#94Faster R-CNN (ResNet-50-FPN, GRoIE)
47.8
APL· 2020-04-28
A novel Region of Interest Extraction Layer for Instance Segmentation Code
#95GHM-C + GHM-R (RetinaNet-FPN-ResNet-50, M=30)
46.7
APL· 2018-11-13
Gradient Harmonized Single-stage Detector Code