AI Paper Atlas

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

Multi-Adversarial Faster-RCNN for Unrestricted Object Detection

Zhenwei He, Lei Zhang

ICCV 2019 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 380 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 hierarchical domain feature alignment to proposal alignment (ablation)MAF · Cityscapes -> Foggy CityscapesmAP: 28.5 -> 34.0Improved
add proposal feature alignment to domain feature alignment (ablation)MAF · Cityscapes -> Foggy CityscapesmAP: 33.6 -> 34.0Improved
add domain classifiers on blocks 3 and 4 to a single block-5 classifier (ablation)MAF* · Cityscapes -> Foggy CityscapesmAP: 30.9 -> 34.0Improved
add WGRL in place of standard GRL (ablation)MAF* · Cityscapes -> Foggy CityscapesmAP: 30.3 -> 30.9Improved
add aggregation of detection results with proposal feature (ablation)MAF* · Cityscapes -> Foggy CityscapesmAP: 30.2 -> 30.9Improved
add hierarchical domain feature alignment (ablation)MAF · SIM10k -> CityscapesAP: 40.1 -> 41.1Improved
add proposal feature alignment to domain feature alignment (ablation)MAF · SIM10k -> CityscapesAP: 40.7 -> 41.1Improved
add hierarchical domain feature alignment (ablation)MAF · Cityscapes -> KITTIAP: 69.9 -> 72.1Improved
add proposal feature alignment to domain feature alignment (ablation)MAF · KITTI -> CityscapesAP: 39.7 -> 41.0Improved
use MAF instead of DAFMAF · Cityscapes -> Foggy CityscapesmAP: 27.6 -> 34.0Improved
use MAF instead of DAFMAF · Cityscapes -> KITTIAP: 64.1 -> 72.1Improved
use MAF instead of DAFMAF · SIM10k -> CityscapesAP: 39.0 -> 41.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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