A Crowd Counting Framework Combining with Crowd Location

Figure 8

Results of sample images from Beijing-BRT (1^st row), Mall (2^nd row), Shanghai Part A (3^rd row), Shanghai Part B (4^th row), and UCF_CC_50 (5^th row). The original pictures in the first column are sampled from different datasets. The pictures in the second column represent the corresponding ground-truth peak confidence map. The pictures in the third column represent the corresponding estimated peak confidence map. The pictures in the fourth column represents the ground-truth (indicating in red) and estimated (indicating in blue) position.