Illuminating Pedestrians via Simultaneous Detection and Segmentation
Citations
420 citations
386 citations
Cites methods from "Illuminating Pedestrians via Simult..."
...We therefore ablate using bins of [4, 8, 16, 32] as described in Tab....
[...]
286 citations
244 citations
References
123,388 citations
55,235 citations
"Illuminating Pedestrians via Simult..." refers methods in this paper
...Thus, we choose to construct a separate network using VGG-16....
[...]
...For a modest improvement to efficiency, we remove the pool5 layer from the VGG-16 architecture then adjust the input size to 112 × 112 to keep the fully-connected layers intact....
[...]
...The RPN architecture uses conv1-5 from VGG-16 [24] as the backbone....
[...]
...We initialize conv1-5 from a VGG-16 model pretrained on ImageNet [7], and all remaining layers randomly....
[...]
49,914 citations
49,639 citations
"Illuminating Pedestrians via Simult..." refers methods in this paper
...We initialize conv1-5 from a VGG-16 model pretrained on ImageNet [7], and all remaining layers randomly....
[...]
40,257 citations