AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Deep Mixture of Linear Inverse Regressions Applied to Head-Pose Estimation

Stephane Lathuiliere, Remi Juge, Pablo Mesejo, Rafael Munoz-Salinas, Radu Horaud

CVPR 2017 · Learning theory and algorithms › Prediction and decision losses · approach: Theory of neural network sensitivity · 58 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Replace the Euclidean-loss output layer with a jointly trained Gaussian mixture of linear inverse regressionsVGG-16 head-pose estimator · Biwimean absolute error (degrees): 4.34 -> 3.62Improved
Use inverse regression instead of SVRHOG features · Biwimean absolute error (degrees): 6.66 -> 5.62Improved
Fine-tune the network using inverse regressionVGG-16 head-pose estimator · Biwimean absolute error (degrees): 17.60 -> 3.62Improved
Use K=1 instead of K=2 Gaussian components (ablation)VGG-IR-FT · Biwipitch MAE (degrees): 4.68 -> 5.12Hurt
Use K=1 instead of K=2 Gaussian components (ablation)VGG-IR-FT · S2G and SSVpitch, yaw, and roll MAE (degrees): S2G: 2.87/3.54/1.65 vs 2.53/3.36/1.97; SSV: 3.12/3.75/2.01 vs 2.86/3.83/2.04Mixed
Use K=5 instead of K=2 Gaussian components (ablation)VGG-IR-FT · Biwipitch, yaw, and roll MAE (degrees): 4.68/3.12/3.07 -> 5.55/3.74/4.01Hurt
Use K=10 instead of K=2 Gaussian components (ablation)VGG-IR-FT · Biwipitch, yaw, and roll MAE (degrees): 4.68/3.12/3.07 -> 5.93/3.45/3.99Hurt
Reduce the training set from 100% to 20%VGG-IR-FT · Biwimean absolute error (degrees): 3.62 -> 7.55Hurt
Use only RGB rather than additional or depth informationVGG-IR-FT · Biwimean absolute error (degrees): 4.23 -> 3.62Improved
Use K=1 by default rather than a larger mixture (ablation)VGG-IR-FT · Biwi, S2G, and SSVpitch, yaw, and roll MAE (degrees): Biwi K=1: 5.12/3.45/3.39; K=2: 4.68/3.12/3.07; K=5: 5.55/3.74/4.01; K=10: 5.93/3.45/3.99Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy