Name
Abstract | ||
|---|---|---|
Current CNN-based solutions to salient object detection (SOD) mainly rely on the optimization of cross-entropy loss (CELoss). Then the quality of detected saliency maps is often evaluated in terms of F-measure. In this paper, we investigate an interesting issue: can we consistently use the F-measure formulation in both training and evaluation for SOD? By reformulating the standard F-measure we propose the relaxed F-measure which is differentiable w.r.t the posterior and can be easily appended to the back of CNNs as the loss function. Compared to the conventional cross-entropy loss of which the gradients decrease dramatically in the saturated area, our loss function, named FLoss, holds considerable gradients even when the activation approaches the target. Consequently, the FLoss can continuously force the network to produce polarized activations. Comprehensive benchmarks on several popular datasets show that FLoss outperforms the state- of-the-arts with a considerable margin. More specifically, due to the polarized predictions, our method is able to obtain high quality saliency maps without carefully tuning the optimal threshold, showing significant advantages in real world applications. |
Year | Venue | Field |
|---|---|---|
2018 | International Conference on Computer Vision | Salient object detection,Pattern recognition,Computer science,Salience (neuroscience),Differentiable function,Artificial intelligence |
DocType | Volume | Citations |
Journal | abs/1805.07567 | 1 |
PageRank | References | Authors |
0.35 | 9 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Kai Zhao | 1 | 1 | 0.35 |
| Shanghua Gao | 2 | 110 | 7.00 |
| Qi-Bin Hou | 3 | 234 | 14.16 |
| Dandan Li | 4 | 27 | 7.99 |
| Ming-Ming Cheng | 5 | 1914 | 82.32 |