Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Energy-Based Models for Deep Probabilistic Regression

About

While deep learning-based classification is generally tackled using standardized approaches, a wide variety of techniques are employed for regression. In computer vision, one particularly popular such technique is that of confidence-based regression, which entails predicting a confidence value for each input-target pair (x,y). While this approach has demonstrated impressive results, it requires important task-dependent design choices, and the predicted confidences lack a natural probabilistic meaning. We address these issues by proposing a general and conceptually simple regression method with a clear probabilistic interpretation. In our proposed approach, we create an energy-based model of the conditional target density p(y|x), using a deep neural network to predict the un-normalized density from (x,y). This model of p(y|x) is trained by directly minimizing the associated negative log-likelihood, approximated using Monte Carlo sampling. We perform comprehensive experiments on four computer vision regression tasks. Our approach outperforms direct regression, as well as other probabilistic and confidence-based methods. Notably, our model achieves a 2.2% AP improvement over Faster-RCNN for object detection on the COCO dataset, and sets a new state-of-the-art on visual tracking when applied for bounding box estimation. In contrast to confidence-based methods, our approach is also shown to be directly applicable to more general tasks such as age and head-pose estimation. Code is available at https://github.com/fregu856/ebms_regression.

Fredrik K. Gustafsson, Martin Danelljan, Goutam Bhat, Thomas B. Sch\"on• 2019

Related benchmarks

TaskDatasetResultRank
Object DetectionCOCO v2017 (test-dev)
mAP39.4
499
Visual Object TrackingTrackingNet (test)
Normalized Precision (Pnorm)80.1
460
Visual TrackingUAV123
AUC67.2
41
Age EstimationUTKFace
MAE4.65
13
Age EstimationUTKFace 34 (test)
MAE4.65
6
Head Pose EstimationBiwi 36 (test)
Yaw Error2.67
6
Age EstimationUTK (test)
MAE4.65
4
Showing 7 of 7 rows

Other info

Code

Follow for update