Papers › Deep Label Distribution Learning with Label Ambiguity

Deep Label Distribution Learning with Label Ambiguity

6 Nov 2016arXiv:1611.01731archive 2025-07-28

Bin-Bin Gao, Chao Xing, Chen-Wei Xie, Jianxin Wu, Xin Geng

Convolutional Neural Networks (ConvNets) have achieved excellent recognition performance in various visual recognition tasks. A large labeled training set is one of the most important factors for its success. However, it is difficult to collect sufficient training images with precise labels in some domains such as apparent age estimation, head pose estimation, multi-label classification and semantic segmentation. Fortunately, there is ambiguous information among labels, which makes these tasks different from traditional classification. Based on this observation, we convert the label of each image into a discrete label distribution, and learn the label distribution by minimizing a Kullback-Leibler divergence between the predicted and ground-truth label distributions using deep ConvNets. The proposed DLDL (Deep Label Distribution Learning) method effectively utilizes the label ambiguity in both feature learning and classifier learning, which help prevent the network from over-fitting even when the training set is small. Experimental results show that the proposed approach produces significantly better results than state-of-the-art methods for age estimation and head pose estimation. At the same time, it also improves recognition performance for multi-label classification and semantic segmentation tasks.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

gaobb/DLDL official report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Age EstimationClassificationGeneral ClassificationHead Pose EstimationMUlTI-LABEL-ClASSIFICATIONMulti-Label ClassificationPose EstimationSemantic Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Age Estimation ChaLearn 2015 DLDL+VGG-Face MAE 3.51 #6 of 7 Archive leaderboard report
Age Estimation ChaLearn 2015 DLDL+VGG-Face e-error 0.31 #6 of 7 Archive leaderboard report
Age Estimation MORPH Album2 DLDL+VGG-Face MAE 2.42±0.01 #5 of 9 Archive leaderboard report
Age Estimation MORPH Album2 DLDL+VGG-Face (KL, Max)3 MAE 2.42 #6 of 9 Archive leaderboard report
Head Pose Estimation AFLW DLDL (KL) MAE 9.78 #6 of 6 Archive leaderboard report
Head Pose Estimation BJUT-3D Ours DLDL (KL) MAE 0.09 #1 of 1 Archive leaderboard report
Head Pose Estimation Pointing'04 Ours DLDL (KL) MAE 4.64 #1 of 1 Archive leaderboard report
Multi-Label Classification PASCAL VOC 2007 Ours PF-DLDL mAP 93.4 #15 of 17 Archive leaderboard report
Multi-Label Classification PASCAL VOC 2012 Ours PF-DLDL mAP 92.4 #3 of 3 Archive leaderboard report
Semantic Segmentation PASCAL VOC 2011 DLDL-8s+CRF Mean IoU 67.6 #1 of 1 Archive leaderboard report
Semantic Segmentation PASCAL VOC 2012 DLDL-8s+CRF Mean IoU 67.1 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections