AI Paper Atlas

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

Recurrent Image Annotation With Explicit Inter-Label Dependencies

Ayushi Dutta, Yashaswi Verma, C.V. Jawahar

ECCV 2020 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 9 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
replace frequent-first label ordering with multi-order training (mask previously chosen label)CNN-RNN · MS-COCO valper-label F1: 62.63 -> 70.13Improved
replace frequent-first label ordering with multi-order trainingCNN-RNN · MS-COCO valper-image F1: 68.13 -> 80.11Improved
replace rare-first ordering with multi-order trainingCNN-RNN · MS-COCO valper-label F1: 63.43 -> 70.13Improved
add an RNN on top of the CNN (frequent-first order)CNN · MS-COCO valper-image F1: 63.30 -> 68.13Improved
use rare-first instead of frequent-first label orderCNN-RNN · MS-COCO valper-image F1: 68.13 -> 67.64Hurt
use rare-first instead of frequent-first label orderCNN-RNN · MS-COCO valper-label F1: 62.63 -> 63.43Improved
replace candidate-label-pool order-free training with multi-order trainingOrder-free RNN · MS-COCO valper-image F1: 67.70 -> 80.11Improved
replace rare-to-frequent training with multi-order trainingSR-CNN-RNN · MS-COCO valper-image F1: 72.47 -> 80.11Improved
replace rare-to-frequent training with multi-order trainingSR-CNN-RNN · MS-COCO valtop-1 accuracy: 81.44 -> 93.49Improved
replace rare-to-frequent training with multi-order trainingSR-CNN-RNN · NUS-WIDE testtop-1 accuracy: 68.06 -> 84.05Improved
replace PLA with multi-order trainingPLA · NUS-WIDE testper-image F1: 72.37 -> 74.74Improved
compare Multi-order RNN with ACfsMulti-order RNN · MS-COCO valper-label F1: 72.20 -> 70.13Hurt

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