| Radhakrishna Achanta†, Sheila Hemami†, Francisco Estrada†, and Sabine Süsstrunk†
This paper introduces a novel method for salient region detection in images, which outputs full-resolution saliency maps with well-defined boundaries of salient objects. The method retains more frequency content from the original image compared to other techniques, making it more effective in preserving the details of salient regions. The approach is based on the biological concept of center-surround contrast but is not limited to any specific biological model. The paper compares the proposed method with five state-of-the-art saliency detection methods using a frequency domain analysis, ground truth, and a salient object segmentation application. The results show that the proposed method outperforms the other algorithms in terms of both precision and recall, demonstrating its superior performance in various applications such as object segmentation, adaptive compression, and object recognition. The method is computationally efficient and simple to implement, making it a promising solution for salient region detection.This paper introduces a novel method for salient region detection in images, which outputs full-resolution saliency maps with well-defined boundaries of salient objects. The method retains more frequency content from the original image compared to other techniques, making it more effective in preserving the details of salient regions. The approach is based on the biological concept of center-surround contrast but is not limited to any specific biological model. The paper compares the proposed method with five state-of-the-art saliency detection methods using a frequency domain analysis, ground truth, and a salient object segmentation application. The results show that the proposed method outperforms the other algorithms in terms of both precision and recall, demonstrating its superior performance in various applications such as object segmentation, adaptive compression, and object recognition. The method is computationally efficient and simple to implement, making it a promising solution for salient region detection.