AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

SMART: Simultaneous Multi-Agent Recurrent Trajectory Prediction

Sriram N N, Buyu Liu, Francesco Pittaluga, Manmohan Chandraker

ECCV 2020 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Interactive multi-agent trajectory prediction · 30 citations (OpenAlex)

Read the paper

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 pretraining on simulated P-ArgoT before fine-tuning on ArgoTSMART (c_best) · ArgoT test 5.0sADE: 2.45 -> 2.02Improved
add pretraining on simulated P-ArgoT before fine-tuning on ArgoTSMART (c_random) · ArgoT test 5.0sADE: 2.88 -> 2.39Improved
add pretraining on simulated P-ArgoT before fine-tuning on ArgoTS-GAN · ArgoT test 5.0sADE: 4.06 -> 4.74Hurt
add pretraining on simulated P-ArgoT before fine-tuning on ArgoTMATF GAN · ArgoT test 5.0sADE: 4.90 -> 5.07Hurt
add pretraining on simulated P-ArgoTS-GAN · ArgoT test without straight trajectories 5.0sADE: 5.47 -> 4.73Improved
add pretraining on simulated P-ArgoTMATF GAN · ArgoT test without straight trajectories 5.0sADE: 5.98 -> 5.61Improved
train on simulated P-ArgoT instead of ArgoT, test on KITTISMART (c_best) · KITTI 5.0sADE: 5.65 -> 5.32Improved
sample trajectory labels equally and take best instead of random labelsSMART · ArgoF val 3.0sADE: 1.65 -> 1.44Improved
sample trajectory labels equally and take best instead of random labelsSMART · ArgoF val 3.0sFDE: 3.00 -> 2.47Improved
use single-pass ConvLSTM over all agents instead of per-agent prediction as agents growSMART vs S-GAN · Argoverse runtimeseconds per sample: .070 -> .069 for SMART vs .024 -> .108 for S-GAN from 1 to 10 agentsImproved
simulate diverse maneuvers instead of using real tracklets onlydataset · KITTI Y Wasserstein meandiversity: 0.14 -> 2.13Improved
simulate diverse maneuvers instead of using real tracklets onlydataset · ArgoT second-derivative-x Wasserstein meandiversity: 5.98 -> 17.5Improved

Later papers that compared with this method

5 later accepted papers report SMART* as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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