AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Deep Hash Distillation for Image Retrieval

Young Kyun Jang, Geonmo Gu, Byungsoo Ko, Isaac Kang, Nam Ik Cho

ECCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 53 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 self-distilled hashing (ablation)HashNet (AlexNet) · ImageNet 16-bitmAP: 0.337 -> 0.501Improved
add self-distilled hashing (ablation)DCH (AlexNet) · ImageNet 16-bitmAP: 0.571 -> 0.640Improved
add self-distilled hashing (ablation)CSQ (AlexNet) · ImageNet 16-bitmAP: 0.569 -> 0.634Improved
add self-distilled hashing (ablation)DPN (AlexNet) · NUS-WIDE 16-bitmAP: 0.753 -> 0.757No clear effect
add self-distilled hashing (ablation)hash proxy loss (AlexNet) · ImageNet 16-bitmAP: 0.574 -> 0.642Improved
add self-distilled hashing (ablation)hash proxy loss (AlexNet) · MS COCO 16-bitmAP: 0.706 -> 0.725Improved
add BCE-based quantization loss to hash proxy loss (ablation)DHD without SdH (AlexNet) · ImageNet 16-bitmAP: 0.574 -> 0.583Improved
add BCE-based quantization loss to hash proxy loss (ablation)DHD with SdH (AlexNet) · ImageNet 16-bitmAP: 0.642 -> 0.657Improved
add self-distilled hashing (ablation)DHD (ResNet50) · ImageNet, no deformationmAP: 0.871 -> 0.891Improved
add self-distilled hashing (ablation)DHD (ResNet50) · ImageNet unseen zoom inmAP: 0.552 -> 0.658Improved
add self-distilled hashing (ablation)DHD (ResNet50) · ImageNet unseen Gaussian noisemAP: 0.673 -> 0.768Improved
add self-distilled hashing (ablation)DHD (ResNet50) · ImageNet unseen zoom outmAP: 0.805 -> 0.816Improved

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