AI Paper Atlas

Research map / Detection and segmentation / CNN features

Scaling Object Detection by Transferring Classification Weights

Jason Kuen, Federico Perazzi, Zhe Lin, Jianming Zhang, Yap-Peng Tan

ICCV 2019 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 19 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 input normalization (ablation)WTN · VG-200AR@100: 41.91 -> 54.19Improved
add group normalization (ablation)WTN · VG-200AR@100: 41.91 -> 58.55Improved
add both input and group normalization (WTN+) (ablation)WTN · OI-500AP50: 52.87 -> 58.82Improved
add both input and group normalization (WTN+) (ablation)WTN · OI-57AP50: 34.94 -> 39.28Improved
add both input and group normalization (WTN+) (ablation)WTN · VG-200AR@100: 41.91 -> 65.60Improved
add autoencoder reconstruction loss (AE-WTN)WTN+ · OI-500AP50: 58.82 -> 59.59Improved
add autoencoder reconstruction loss (AE-WTN)WTN+ · OI-57AP50: 39.28 -> 41.07Improved
add autoencoder reconstruction loss (AE-WTN)WTN+ · VG-200AR@100: 65.60 -> 66.75Improved
add activity regularizer instead of reconstruction loss (ablation)WTN+ · VG-200AR@100: 65.60 -> 36.26Hurt
add 5x weight decay (ablation)WTN+ · OI-57AP50: 39.28 -> 40.79Improved
add Dropout (ablation)WTN+ · OI-57AP50: 39.28 -> 40.09Improved
halve hidden channels (reduced capacity) (ablation)WTN+ · OI-57AP50: 39.28 -> 37.81Hurt

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