AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

An Uncertain Future: Forecasting from Static Images Using Variational Autoencoders

Jacob Walker, Carl Doersch, Abhinav Gupta, Martial Hebert

ECCV 2016 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Disentangled and discrete latent models · 476 citations (OpenAlex)

Read the paperCheck 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
add a latent-variable VAE to the direct regressortrajectory predictor · THUMOS 2015 trajectory predictionnegative log likelihood: 11463 -> 11082Improved
use VAE trajectory model instead of optical flow extrapolationtrajectory predictor · THUMOS 2015 trajectory predictionnegative log likelihood: 11734 -> 11082Improved
sample multiple predictions from the VAE instead of regressor outputtrajectory predictor · THUMOS 2015 minimum Euclidean distanceaverage minimum Euclidean distanceMixed
pretrain on trajectory prediction with calibration instead of training from scratchFast-RCNN AlexNet-style · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 40.6 -> 47.3Improved
pretrain on trajectory prediction instead of kmeans initializationFast-RCNN · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 46.1 -> 47.3Improved
use trajectory prediction pretraining instead of relative position on ImageNetFast-RCNN · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 52.3 -> 47.3Hurt
pretrain on trajectory prediction instead of egomotionFast-RCNN · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 44.9 -> 47.3Improved
add between-layer scale calibration (ablation)video embedding pretraining (8M frames) · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 47.5 -> 42.2Hurt
increase video embedding pretraining from 1.5M to 8M framesvideo embedding baseline · PASCAL VOC 2012 test detection (Fast-RCNN)mAP: 46.2 -> 47.5Improved
use trajectory pretraining instead of relative positionFast-RCNN · PASCAL VOC 2012 test detection, person class (Fast-RCNN)AP: 57.5 -> 58.4Improved
pretrain on trajectory prediction instead of relative positionFast-RCNN · PASCAL VOC 2012 test detection, chair class (Fast-RCNN)AP: 27.0 -> 21.2Hurt

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