Timezone: »
In crowd counting, each training image contains multiple people, where each person is annotated by a dot. Existing crowd counting methods need to use a Gaussian to smooth each annotated dot or to estimate the likelihood of every pixel given the annotated point. In this paper, we show that imposing Gaussians to annotations hurts generalization performance. Instead, we propose to use Distribution Matching for crowd COUNTing (DM-Count). In DM-Count, we use Optimal Transport (OT) to measure the similarity between the normalized predicted density map and the normalized ground truth density map. To stabilize OT computation, we include a Total Variation loss in our model. We show that the generalization error bound of DM-Count is tighter than that of the Gaussian smoothed methods. In terms of Mean Absolute Error, DM-Count outperforms the previous state-of-the-art methods by a large margin on two large-scale counting datasets, UCF-QNRF and NWPU, and achieves the state-of-the-art results on the ShanghaiTech and UCF-CC50 datasets. DM-Count reduced the error of the state-of-the-art published result by approximately 16%. Code is available at https://github.com/cvlab-stonybrook/DM-Count.
Author Information
Boyu Wang (Stony Brook University)
Huidong Liu (Stony Brook University)
Dimitris Samaras (Stony Brook University)
Minh Hoai Nguyen (Stony Brook University)
Related Events (a corresponding poster, oral, or spotlight)
-
2020 Spotlight: Distribution Matching for Crowd Counting »
Thu. Dec 10th 03:00 -- 03:10 AM Room Orals & Spotlights: Vision Applications
More from the Same Authors
-
2022 : Learning Probabilistic Topological Representations Using Discrete Morse Theory »
Xiaoling Hu · Dimitris Samaras · Chao Chen -
2022 : Learning Probabilistic Topological Representations Using Discrete Morse Theory »
Xiaoling Hu · Dimitris Samaras · Chao Chen -
2022 Poster: Diffusion Models as Plug-and-Play Priors »
Alexandros Graikos · Nikolay Malkin · Nebojsa Jojic · Dimitris Samaras -
2020 Poster: Detecting Hands and Recognizing Physical Contact in the Wild »
Supreeth Narasimhaswamy · Trung Nguyen · Minh Hoai Nguyen -
2019 Poster: Topology-Preserving Deep Image Segmentation »
Xiaoling Hu · Fuxin Li · Dimitris Samaras · Chao Chen -
2018 Poster: Sequence-to-Segment Networks for Segment Detection »
Zijun Wei · Boyu Wang · Minh Hoai Nguyen · Jianming Zhang · Zhe Lin · Xiaohui Shen · Radomir Mech · Dimitris Samaras -
2016 Poster: Learned Region Sparsity and Diversity Also Predicts Visual Attention »
Zijun Wei · Hossein Adeli · Minh Hoai Nguyen · Greg Zelinsky · Dimitris Samaras -
2013 Poster: Modeling Clutter Perception using Parametric Proto-object Partitioning »
Chen-Ping Yu · Wen-Yu Hua · Dimitris Samaras · Greg Zelinsky