Gaussian process regression (GPR) is a non-parametric Bayesian method that models the relationship between input features and continuous output variables by placing a prior distribution over functions, allowing predictions to be made along with principled uncertainty estimates. In the context of crowd counting and density estimation, GPR has been employed to map handcrafted or learned feature representations to crowd count values, offering a probabilistic alternative to deterministic regressors. Its relevance to the field lies in its ability to provide confidence intervals alongside count predictions and to perform well in low-data regimes, though it has largely been supplanted by deep CNN-based end-to-end approaches that scale more effectively to large and diverse crowd datasets.
Source Papers
- A survey of recent advances in CNN-based single image crowd counting and density estimation ↗ — A survey of recent advances in CNN-based single image crowd
- Radio Radiance Field: The New Frontier of Spatial Wireless Channel Representation ↗ — Radio Radiance Field: The New Frontier of Spatial Wireless C
- Recent trends in crowd analysis: A review ↗ — Recent trends in crowd analysis: A review