AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Rethinking 360deg Image Visual Attention Modelling With Unsupervised Learning.

Yasser Abdelaziz Dahou Djilali, Tarun Krishna, Kevin McGuinness, Noel E. O’Connor

ICCV 2021 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 1 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-attention to the contrastive encoder (ResNet) (ablation)ResNet contrastive encoder + decoder · VR-EyeTracking+SitzmanKLD: 8.574 -> 7.110Improved
add self-attention to the contrastive encoder (VGG, lambda 0.7) (ablation)VGG contrastive encoder + decoder · VR-EyeTracking+SitzmanKLD: 7.965 -> 7.583Improved
add self-attention to the contrastive encoder (VGG, lambda 0.5) (ablation)VGG contrastive encoder + decoder · VR-EyeTracking+SitzmanKLD: 7.540 -> 7.648Hurt
add self-attention to the contrastive encoder (VGG, lambda 0.9) (ablation)VGG contrastive encoder + decoder · VR-EyeTracking+SitzmanKLD: 7.875 -> 7.648Improved
add self-attention to the contrastive encoder (ResNet) (ablation)ResNet contrastive encoder + decoder · Salient360!KLD: 1.039 -> 0.849Improved
increase lambda from 0.5 to 0.7 (more weight on local NCE) (ablation)VGG with self-attention · VR-EyeTracking+SitzmanKLD: 7.648 -> 7.583Improved
increase lambda from 0.7 to 0.9 (ablation)VGG with self-attention · VR-EyeTracking+SitzmanKLD: 7.583 -> 7.648Hurt
add sphere-to-plane projections (top-to-front, bottom-to-front) as contrastive views (ablation)ResNet contrastive encoder · Salient360!KLD: 0.999 -> 0.849Improved
add projections as contrastive views (ablation)ResNet contrastive encoder · Salient360!CC: 0.479 -> 0.584Improved
add projections as contrastive views (ablation)ResNet contrastive encoder · Salient360!AUC-J: 0.736 -> 0.769Improved
use frozen unsupervised 360 encoder instead of 2D SalGAN saliency modelsaliency prediction · VR-EyeTracking+SitzmanKLD: 10.195 -> 7.110Improved
use single ERP stream ResNet model instead of ATSAL360 saliency inference · Salient360! 4K imagesruntime per image (s): 0.230 -> 0.025Improved

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