Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/YOLOv3

YOLOv3

Reported on 75 benchmarks across 5 tasks · 4 papers · 10 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology60 results

3DonCOCO (Common Objects in Context)
FPS (V100, b=1)· 2022-03-30
123
best: 779 (YOLOv6-N)
SOTA
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
2D ClassificationonCOCO (Common Objects in Context)
FPS (V100, b=1)· 2022-03-30
123
best: 779 (YOLOv6-N)
SOTA
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
2D Object DetectiononCOCO (Common Objects in Context)
FPS (V100, b=1)· 2022-03-30
123
best: 779 (YOLOv6-N)
SOTA
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
16konCOCO (Common Objects in Context)
FPS (V100, b=1)· 2022-03-30
123
best: 779 (YOLOv6-N)
SOTA
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
3DonCPPE-5
APM· 2021-11-07
28.4
best: 43.4 (Empirical Attention)
SOTA
Natural Adversarial Objects arXiv:2111.04204
2D ClassificationonCPPE-5
APM· 2021-11-07
28.4
best: 43.4 (Empirical Attention)
SOTA
Natural Adversarial Objects arXiv:2111.04204
2D Object DetectiononCPPE-5
APM· 2021-11-07
28.4
best: 43.4 (Empirical Attention)
SOTA
Natural Adversarial Objects arXiv:2111.04204
16konCPPE-5
APM· 2021-11-07
28.4
best: 43.4 (Empirical Attention)
SOTA
Natural Adversarial Objects arXiv:2111.04204
3DonCOCO (Common Objects in Context)
box AP· 2022-03-30
51
best: 59.5 (DEIM-D-FINE-X+)
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
2D ClassificationonCOCO (Common Objects in Context)
box AP· 2022-03-30
51
best: 59.5 (DEIM-D-FINE-X+)
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
2D Object DetectiononCOCO (Common Objects in Context)
box AP· 2022-03-30
51
best: 59.5 (DEIM-D-FINE-X+)
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
16konCOCO (Common Objects in Context)
box AP· 2022-03-30
51
best: 59.5 (DEIM-D-FINE-X+)
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
3DonCPPE-5
AP50· 2021-12-15
79.4
best: 87.3 (Double Heads)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
3DonCPPE-5
AP75· 2021-12-15
35.3
best: 58.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
3DonCPPE-5
APL· 2021-12-15
49
best: 62.6 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
3DonCPPE-5
APS· 2021-12-15
23.1
best: 45.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
3DonCPPE-5
box AP· 2021-12-15
38.5
best: 52.9 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D ClassificationonCPPE-5
AP50· 2021-12-15
79.4
best: 87.3 (Double Heads)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D ClassificationonCPPE-5
AP75· 2021-12-15
35.3
best: 58.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D ClassificationonCPPE-5
APL· 2021-12-15
49
best: 62.6 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D ClassificationonCPPE-5
APS· 2021-12-15
23.1
best: 45.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D ClassificationonCPPE-5
box AP· 2021-12-15
38.5
best: 52.9 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D Object DetectiononCPPE-5
AP50· 2021-12-15
79.4
best: 87.3 (Double Heads)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D Object DetectiononCPPE-5
AP75· 2021-12-15
35.3
best: 58.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D Object DetectiononCPPE-5
APL· 2021-12-15
49
best: 62.6 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D Object DetectiononCPPE-5
APS· 2021-12-15
23.1
best: 45.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
2D Object DetectiononCPPE-5
box AP· 2021-12-15
38.5
best: 52.9 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
16konCPPE-5
AP50· 2021-12-15
79.4
best: 87.3 (Double Heads)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
16konCPPE-5
AP75· 2021-12-15
35.3
best: 58.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
16konCPPE-5
APL· 2021-12-15
49
best: 62.6 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
16konCPPE-5
APS· 2021-12-15
23.1
best: 45.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
16konCPPE-5
box AP· 2021-12-15
38.5
best: 52.9 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
3DonNAO
mAP· 2021-11-07
10
best: 15.2 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
3DonNAO
mAP w/o OOD· 2021-11-07
17.5
best: 29.6 (EfficientDet-D4)
Natural Adversarial Objects arXiv:2111.04204
3DonNAO
mAR· 2021-11-07
28.4
best: 43.8 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
2D ClassificationonNAO
mAP· 2021-11-07
10
best: 15.2 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
2D ClassificationonNAO
mAP w/o OOD· 2021-11-07
17.5
best: 29.6 (EfficientDet-D4)
Natural Adversarial Objects arXiv:2111.04204
2D ClassificationonNAO
mAR· 2021-11-07
28.4
best: 43.8 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
2D Object DetectiononNAO
mAP· 2021-11-07
10
best: 15.2 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
2D Object DetectiononNAO
mAP w/o OOD· 2021-11-07
17.5
best: 29.6 (EfficientDet-D4)
Natural Adversarial Objects arXiv:2111.04204
2D Object DetectiononNAO
mAR· 2021-11-07
28.4
best: 43.8 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
16konNAO
mAP· 2021-11-07
10
best: 15.2 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
16konNAO
mAP w/o OOD· 2021-11-07
17.5
best: 29.6 (EfficientDet-D4)
Natural Adversarial Objects arXiv:2111.04204
16konNAO
mAR· 2021-11-07
28.4
best: 43.8 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
3DonImageNet VID
MAP · 2020-09-23
68.6
best: 93.2 (YOLOV++)
Robust and efficient post-processing for video object detection arXiv:2009.11050
2D ClassificationonImageNet VID
MAP · 2020-09-23
68.6
best: 93.2 (YOLOV++)
Robust and efficient post-processing for video object detection arXiv:2009.11050
2D Object DetectiononImageNet VID
MAP · 2020-09-23
68.6
best: 93.2 (YOLOV++)
Robust and efficient post-processing for video object detection arXiv:2009.11050
16konImageNet VID
MAP · 2020-09-23
68.6
best: 93.2 (YOLOV++)
Robust and efficient post-processing for video object detection arXiv:2009.11050
3DonNII-CU MAPD
AP@0.5
92.4
3DonNII-CU MAPD
AP@0.75
44.5
3DonNII-CU MAPD
mAP@0.5:0.95
48.3
2D ClassificationonNII-CU MAPD
AP@0.5
92.4
2D ClassificationonNII-CU MAPD
AP@0.75
44.5
2D ClassificationonNII-CU MAPD
mAP@0.5:0.95
48.3
2D Object DetectiononNII-CU MAPD
AP@0.5
92.4
2D Object DetectiononNII-CU MAPD
AP@0.75
44.5
2D Object DetectiononNII-CU MAPD
mAP@0.5:0.95
48.3
16konNII-CU MAPD
AP@0.5
92.4
16konNII-CU MAPD
AP@0.75
44.5
16konNII-CU MAPD
mAP@0.5:0.95
48.3

Computer Vision15 results

Object DetectiononCOCO (Common Objects in Context)
FPS (V100, b=1)· 2022-03-30
123
best: 779 (YOLOv6-N)
SOTA
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
Object DetectiononCPPE-5
APM· 2021-11-07
28.4
best: 43.4 (Empirical Attention)
SOTA
Natural Adversarial Objects arXiv:2111.04204
Object DetectiononCOCO (Common Objects in Context)
box AP· 2022-03-30
51
best: 59.5 (DEIM-D-FINE-X+)
PP-YOLOE: An evolved version of YOLO arXiv:2203.16250
Object DetectiononCPPE-5
AP50· 2021-12-15
79.4
best: 87.3 (Double Heads)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
Object DetectiononCPPE-5
AP75· 2021-12-15
35.3
best: 58.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
Object DetectiononCPPE-5
APL· 2021-12-15
49
best: 62.6 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
Object DetectiononCPPE-5
APS· 2021-12-15
23.1
best: 45.8 (Localization Distillation)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
Object DetectiononCPPE-5
box AP· 2021-12-15
38.5
best: 52.9 (TridentNet)
CPPE-5: Medical Personal Protective Equipment Dataset arXiv:2112.09569
Object DetectiononNAO
mAP· 2021-11-07
10
best: 15.2 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
Object DetectiononNAO
mAP w/o OOD· 2021-11-07
17.5
best: 29.6 (EfficientDet-D4)
Natural Adversarial Objects arXiv:2111.04204
Object DetectiononNAO
mAR· 2021-11-07
28.4
best: 43.8 (Mask RCNN R50)
Natural Adversarial Objects arXiv:2111.04204
Object DetectiononImageNet VID
MAP · 2020-09-23
68.6
best: 93.2 (YOLOV++)
Robust and efficient post-processing for video object detection arXiv:2009.11050
Object DetectiononNII-CU MAPD
AP@0.5
92.4
Object DetectiononNII-CU MAPD
AP@0.75
44.5
Object DetectiononNII-CU MAPD
mAP@0.5:0.95
48.3