Neural Posterior Estimation with Autoregressive Tiling for Detecting Objects in Astronomical Images
This addresses the challenge of small-object detection in astronomy for upcoming petabyte-scale surveys, representing an incremental advance with a novel method for a known bottleneck.
The paper tackles the problem of detecting faint and overlapping astronomical objects in large-scale sky surveys by proposing an amortized variational inference method with autoregressive tiling, achieving state-of-the-art performance on Sloan Digital Sky Survey images and improving posterior calibration.
Upcoming astronomical surveys will produce petabytes of high-resolution images of the night sky, providing information about billions of stars and galaxies. Detecting and characterizing the astronomical objects in these images is a fundamental task in astronomy -- and a challenging one, as most of these objects are faint and many visually overlap with other objects. We propose an amortized variational inference procedure to solve this instance of small-object detection. Our key innovation is a family of spatially autoregressive variational distributions that partition and order the latent space according to a $K$-color checkerboard pattern. By construction, the conditional independencies of this variational family mirror those of the posterior distribution. We fit the variational distribution, which is parameterized by a convolutional neural network, using neural posterior estimation (NPE) to minimize an expectation of the forward KL divergence. Using images from the Sloan Digital Sky Survey, our method achieves state-of-the-art performance. We further demonstrate that the proposed autoregressive structure greatly improves posterior calibration.