Codesota · Benchmark · Pascal VOC 2012Home/Leaderboards/Vision & Documents/Object Detection/Pascal VOC 2012
Unknown

Pascal VOC 2012.

11,530 images with 27,450 ROI annotated objects and 6,929 segmentations. Classic object detection benchmark.

Paper ↗Leaderboard ↓
§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Miou

Miou is the reported evaluation metric for Pascal VOC 2012. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Miouverifiedpapervendorcommunityunverified

MAP Coco Pretrain

MAP Coco Pretrain is the reported evaluation metric for Pascal VOC 2012. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for MAP Coco Pretrainverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01SSD512 (VGG-16)
SSD512 VGG-16 with COCO pre-training. 80.0 mAP on PASCAL VOC 2012 test — best result in SSD paper. Trained on VOC07+12+COCO. Table 4, arxiv:1512.02325.
verified802015Paper ↗Looks wrong?
02SSD300 (VGG-16)
SSD300 VGG-16 with COCO pre-training. 77.5 mAP on PASCAL VOC 2012 test. Trained on VOC07+12+COCO. Table 4, arxiv:1512.02325.
verified77.52015Paper ↗Looks wrong?
03Faster R-CNN (VGG-16)
Faster R-CNN VGG-16 with COCO pre-training. 75.9 mAP on PASCAL VOC 2012 test. Trained on VOC07+12+COCO. Table 4, arxiv:1506.01497.
verified75.92015Paper ↗Looks wrong?

mAP

MAP is the reported evaluation metric for Pascal VOC 2012. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for mAPverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01SSD512 (VGG-16)
SSD512 VGG-16. 74.9 mAP on PASCAL VOC 2012 test. Trained on VOC07+12. Table 4, arxiv:1512.02325. ECCV 2016.
verified74.92015Paper ↗Looks wrong?
02SSD300 (VGG-16)
SSD300 VGG-16. 72.4 mAP on PASCAL VOC 2012 test. Trained on VOC07+12. Table 4, arxiv:1512.02325. ECCV 2016.
verified72.42015Paper ↗Looks wrong?
03Faster R-CNN (VGG-16)
Faster R-CNN VGG-16. 70.4 mAP on PASCAL VOC 2012 test. Trained on VOC07+12 (07 trainval+test + 12 trainval). Table 4, arxiv:1506.01497. NeurIPS 2015.
verified70.42015Paper ↗Looks wrong?
§ 04 · Submit a result

Add to the leaderboard.

← Back to Object Detection