SOTAVerified

Saliency Prediction

A saliency map is a model that predicts eye fixations on a visual scene. Saliency prediction is informed by the human visual attention mechanism and predicts the possibility of the human eyes to stay in a certain position in the scene.

Papers

Showing 231240 of 268 papers

TitleStatusHype
SalGAN: Visual Saliency Prediction with Generative Adversarial NetworksCode0
Predicting Human Eye Fixations via an LSTM-based Saliency Attentive ModelCode0
Deep, Dense, and Low-Rank Gaussian Conditional Random Fields0
DeepGaze II: Reading fixations from deep features trained on object recognition0
Probabilistic Saliency Estimation0
UberNet: Training a `Universal' Convolutional Neural Network for Low-, Mid-, and High-Level Vision using Diverse Datasets and Limited MemoryCode0
A Deep Multi-Level Network for Saliency PredictionCode0
Deeply-Supervised Recurrent Convolutional Neural Network for Saliency Detection0
Saliency Detection via Combining Region-Level and Pixel-Level Predictions with CNNs0
Spatio-Temporal Saliency Networks for Dynamic Saliency Prediction0
Show:102550
← PrevPage 24 of 27Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TransalnetKL0.87Unverified
2TempSALKL0.71Unverified
3SSwin transformerKL0.65Unverified
4ECT-SALKL0.58Unverified
5SUMKL0.47Unverified
#ModelMetricClaimedVerifiedStatus
1SUMAUC0.88Unverified
2SalNAS-XL + Self-KDAUC0.87Unverified
3TempSALAUC0.87Unverified
4MDS-ViTNetAUC0.87Unverified
5TranSalNetAUC0.87Unverified
#ModelMetricClaimedVerifiedStatus
1SUMAUC-Judd0.91Unverified
2TranSalNetAUC-Judd0.87Unverified
#ModelMetricClaimedVerifiedStatus
1SUMKL0.27Unverified