AI Paper Atlas

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

Modality Translation for Object Detection Adaptation without forgetting prior knowledge

Heitor Rapela Medeiros, Masih Aminbeidokhti, Fidel A Guerrero Pena, David Latortue, Eric Granger, Marco Pedersoli

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 6 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
train a small input translator with detection loss instead of image translation (CycleGAN/CUT/HalluciDet)frozen RGB detector · LLVIP IR test, FCOSAP: 28.00 -> 57.63 vs HalluciDetImproved
use ModTr instead of HalluciDet with Faster R-CNN on LLVIPFaster R-CNN · LLVIP IR testAP: 57.78 -> 57.97No clear effect
use ModTr instead of HalluciDet on FLIRFCOS · FLIR IR testAP: 23.74 -> 35.49Improved
ModTr with frozen detector instead of full fine-tuning on FLIRFCOS · FLIR IR testAP: 27.97 -> 35.49Improved
ModTr instead of full fine-tuning on LLVIPFaster R-CNN · LLVIP IR testAP: 59.62 -> 57.97Hurt
LoRA on last layer of detector instead of ModTrLoRA · LLVIP/FLIRAP: LoRA FCOS LLVIP 47.72 vs ModTr 57.63Hurt
shrink translator from ResNet34 to MobileNetv2 (ablation)ModTr_odot · LLVIP, Faster R-CNNAP: 56.35 -> 56.15No clear effect
shrink translator from ResNet34 to MobileNetv3s (ablation)ModTr_odot · FLIR, Faster R-CNNAP: 37.21 -> 32.06Hurt
fine-tune a single detector jointly on all modalities (1-Detector) instead of ModTrdetector · COCO val, FCOSAP: 38.41 -> 00.33Hurt
use ModTr (frozen detector) rather than N-Detectors for knowledge preservationFCOS, RetinaNet, Faster R-CNN · Average over LLVIP, FLIR, COCOAP: FCOS 41.25 -> 43.84Improved
also fine-tune the detector jointly with ModTrModTr_odot · FLIRAPImproved
add image-quality loss terms to the translator (ablation)ModTr · LLVIP/FLIRAPNo clear effect

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