AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Joint Learning of Object and Action Detectors

Vicky Kalogeiton, Philippe Weinzaepfel, Vittorio Ferrari, Cordelia Schmid

ICCV 2017 · Detection and segmentation › Open-set and few-shot detection · approach: Visual relationship and scene graph detection · 62 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 flow stream to RGB with late score fusion (ablation)RGB detector · A2D object detectionmAP: 63.1 -> 61.6Hurt
add flow stream to RGB with late score fusion (ablation)RGB detector · ILSVRC VID object detectionmAP: 45.2 -> 33.9Hurt
use proposed RoI and feature-level fusion instead of late score fusion (ablation)two-stream detector · A2D object detectionmAP: 61.6 -> 65.3Improved
use proposed fusion instead of late score fusion (ablation)two-stream detector · ILSVRC VID object detectionmAP: 33.9 -> 48.1Improved
train multitask on objects and actions instead of objects onlydetector · A2D objectsmAP: 65.3 -> 68.3Improved
train multitask on objects and actions instead of actions onlydetector · A2D actionsmAP: 56.2 -> 60.0Improved
joint multitask training instead of separate detectors with multiplied scoresdetector · A2D objects+actionsmAP: 43.1 -> 48.9Improved
use Cartesian product labels instead of multitask (ablation)detector · A2D objects+actionsmAP: 48.9 -> 49.2No clear effect
use hierarchical labels instead of multitask (ablation)detector · A2D objects+actionsmAP: 48.9 -> 49.6No clear effect
vary action loss weight over 0.5, 1, 2 (ablation)multitask detector · A2D object-actionmAP: < 0.5% variationNo clear effect
use SharpMask proposals instead of GBH (ablation)segmentation from detections · A2D object segmentationmIoU: 42.7 -> 49.5Improved
use multitask architecture instead of separate networks with multiplied scoresrelationship detector · VRD phrase detectionR@100: 11.9 -> 18.3Improved

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