atlas

Research map / Architectures and efficiency / RNNs and normalization

On Human Motion Prediction Using Recurrent Neural Networks

CVPR 2017 · Architectures and efficiency › RNNs and normalization · approach: Efficient recurrent architectures · 1,024 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 recurrent prediction model with a zero-velocity predictorHuman 3.6M motion prediction · Human 3.6M Average, 400 msmean angle error: 1.20 -> 1.15Hurt
use sampling-based loss instead of feeding ground truth at each decoder step (ablation)single-action model · Human 3.6M Walking, 400 msmean angle error: 1.34 -> 1.20Helped
add residual connections to model motion derivatives (ablation)single-action model with sampling-based loss · Human 3.6M Walking, 400 msmean angle error: 1.20 -> 1.09Helped
train the residual model on multiple actions rather than one action (ablation)Human 3.6M motion prediction · Human 3.6M Walking, 400 msmean angle error: 1.09 -> 0.81Helped
add action labels to the multi-action model (ablation)residual multi-action model · Human 3.6M Walking, 400 msmean angle error: 0.78 -> 0.81Hurt
share encoder and decoder weights (ablation)residual supervised multi-action model · Human 3.6M Walking, 400 msmean angle error: 0.91 -> 0.81Helped
use the zero-velocity predictor instead of SRNNHuman 3.6M motion prediction · Human 3.6M Walking, 400 msmean angle error: 1.30 -> 1.15Helped

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