jacobs-university / iccv3_salient_object

ICCV 3 - Scene Context-Aware Salient Object Detection

ICCV Scene Context-Aware Salient Object Detection

ICCV 3 - Scene Context-Aware Salient Object Detection

This repository implements Scene Context-Aware Object Detection based on this paper : https://openaccess.thecvf.com/content/ICCV2021/papers/Siris_Scene_Context-Aware_Salient_Object_Detection_ICCV_2021_paper.pdf

NOTE : Detectron2 is not originally for supported for Windows so it can give issues while installation.

For Salient Object Detection - The train and test images can be found under SCAS_Dataset. In this repository you can also find Salinet Object Detection done with other methods such as:

1. BASNet - Boundary-Aware Segmentation Network
2. EGNet - Edge Guidance Network for Salient Object Detection
3. PFA-Net - Pyramid Feature Attention Network for Saliency Detection

For all these methods, the respective codes and the train-test data can be found seperately in this repository.

Authors:

Deepali Bohra, Sayali Rajhans, Pushpak Kashid