AI Paper Atlas

Research map / Learning theory and algorithms / Prediction and decision losses

Bayesian Loss for Crowd Count Estimation With Point Supervision

Zhiheng Ma, Xing Wei, Xiaopeng Hong, Yihong Gong

ICCV 2019 · Learning theory and algorithms › Prediction and decision losses · approach: Ranking-based loss optimization · 685 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 pixel-wise Gaussian density loss with Bayesian lossVGG-19 density estimator · UCF-QNRFMAE: 106.8 -> 92.9Improved
add background pixel modelling to Bayesian loss (ablation)Bayesian loss, VGG-19 · UCF-QNRFMAE: 92.9 -> 88.7Improved
replace Gaussian-target loss with Bayesian+ lossVGG-19 · ShanghaiTech AMAE: 68.6 -> 62.8Improved
replace Gaussian-target loss with Bayesian+ lossVGG-19 · ShanghaiTech BMAE: 8.5 -> 7.7Improved
replace Gaussian-target loss with Bayesian+ lossVGG-19 · UCF_CC_50MAE: 251.6 -> 229.3Improved
limit image shorter side to 2048 pixels (ablation)Bayesian+ · UCF-QNRFMAE: 112.6 -> 88.7Improved
limit image shorter side to 2048 pixels (ablation)Baseline · UCF-QNRFMAE: 128.7 -> 106.8Improved
switch backbone from AlexNet to VGG-19 (ablation)Bayesian+ · UCF-QNRFMAE: 116.3 -> 88.7Improved
replace Gaussian-target loss with Bayesian+ lossAlexNet · UCF-QNRFMAE: 130.8 -> 116.3Improved
replace Gaussian-target loss with Bayesian+ loss, cross-datasetVGG-19 trained on UCF-QNRF · ShanghaiTech A (cross-dataset)MAE: 73.4 -> 69.8Improved
vary Gaussian sigma from 0.1 to 32.0 in the loss (ablation)Bayesian loss · UCF-QNRFMAE: less than 98.0No clear effect
vary Gaussian sigma in the target density map (ablation)Baseline · UCF-QNRFMAE: 118.4 to 136.2Hurt

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