Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Overview of the proposed ERPN.

More »

Fig 1 Expand

Fig 2.

The architecture of deconvolutional feature pyramid network.

More »

Fig 2 Expand

Fig 3.

Diagram of the interspersed scales for novel anchor boxes.

More »

Fig 3 Expand

Fig 4.

Comparison between the anchor boxes of RPN and ERPN.

More »

Fig 4 Expand

Fig 5.

The flowchart of optimizing the SVM parameters with PSO.

More »

Fig 5 Expand

Fig 6.

The PSO iteration number.

More »

Fig 6 Expand

Fig 7.

The classification loss function in RPN.

More »

Fig 7 Expand

Fig 8.

The novel coefficients of improved classification loss function in ERPN.

More »

Fig 8 Expand

Table 1.

Data set information.

More »

Table 1 Expand

Table 2.

Parameters for PSO.

More »

Table 2 Expand

Table 3.

Parameters of Fast R-CNN.

More »

Table 3 Expand

Table 4.

Parameters of MR-CNN.

More »

Table 4 Expand

Table 5.

Parameters of ION.

More »

Table 5 Expand

Table 6.

Parameters of ERPN.

More »

Table 6 Expand

Table 7.

Parameters of Faster R-CNN.

More »

Table 7 Expand

Table 8.

Parameters of HyperNet.

More »

Table 8 Expand

Table 9.

Detection results on PASCAL VOC 2007 test set, the best AP of each object category and mAP are bold-faced.

More »

Table 9 Expand

Table 10.

Detection results on PASCAL VOC 2012 test set, the best AP of each object category and mAP are bold-faced.

More »

Table 10 Expand

Fig 9.

Small objects detection results on PASCAL VOC 2007 and VCO 2012 test sets.

More »

Fig 9 Expand

Fig 10.

Recall versus IoU threshold on the PASCAL VOC 2007 test set.

Left: 200 region proposals. Middle: 500 region proposals. Right: 1000 region proposals.

More »

Fig 10 Expand

Table 11.

Detection results on MS COCO test-std.

The best result is bold-faced. A: improved anchor boxes, D: DFPN.

More »

Table 11 Expand

Table 12.

Comparison between SVM and softmax classifiers.

More »

Table 12 Expand

Table 13.

Experiment results for improved loss function.

More »

Table 13 Expand

Fig 11.

The influence for Eta on coefficients of improved classification loss function.

More »

Fig 11 Expand

Table 14.

Detection speed of different methods on the PASCAL VOC 2007 test set.

More »

Table 14 Expand