AI Paper Atlas

Research map / Detection and segmentation / Tracking

FMODetect: Robust Detection of Fast Moving Objects

Denys Rozumnyi, Jiří Matas, Filip Šroubek, Marc Pollefeys, Martin R. Oswald

ICCV 2021 · Detection and segmentation › Tracking · approach: Memory and correlation-based MOT · 15 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 energy-minimization deblurring to refine network trajectories (ablation)FMODetect · TbD datasetTIoU: 0.519 -> 0.715Improved
add energy-minimization deblurring to refine network trajectories (ablation)FMODetect · TbD datasetrecall: 0.97 -> 0.97No clear effect
add energy-minimization deblurring to refine network trajectories (ablation)FMODetect · TbD datasetruntime: 20 fps -> 0.4 fpsHurt
add non-causal post-processing (ablation)FMODetect + deblurring · TbD datasetTIoU: 0.715 -> 0.781Improved
add non-causal post-processing (ablation)FMODetect + deblurring · TbD datasetrecall: 0.97 -> 0.99Improved
add binary FMO predictor output to the matting and fitting network (ablation)FMODetect · TbD datasetfalse positives: removes approximately 70% of false positivesImproved
split deblatting into matting by network then deblurring (removing background first)FMODetect deblurring · TbD datasetdeblurring speed: twice as fast; 10 times faster without HImproved
use FMODetect instead of TbDFMODetect · FMO datasetrecall (%): 41.1 -> 48.8Improved
use FMODetect instead of TbDFMODetect · TbD-3D datasetprecision (%): 98.2 -> 87.9Hurt
use FMODetect instead of TbDFMODetect · TbD-3D datasetrecall (%): 69.8 -> 83.5Improved
use FMODetect instead of TbDFMODetect · YouTube trick shotsrecall (%): 27.3 -> 90.1Improved
use FMODetect instead of TbDFMODetect · Falling objects datasetprecision (%): 92.2 -> 98.9Improved

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