Article · Computers · October 22, 2026
Measure-Based Supervision for Tiny Object Detection in Aerial Imagery
Rashid Nasimov, Jasur Usmonovich Sevinov, Akmalbek Bobomirzaevich Abdusalomov, Yucel Ogurlu, Narziev Nosir, Abbosjon Abdullaev
Abstract
Detection breaks down on objects a few pixels wide for a structural reason: as box side length shrinks, intersection over union (IoU) degenerates toward an indicator of exact pixel coincidence, so a one-pixel error can rewrite both the label assignment and the regression gradient. We avoid this regime by discarding pairwise box matching altogether. A scene is treated instead as an unnormalized positive measure over object centers, log-scales, and categories, and training minimizes a debiased, uncertainty-aware unbalanced optimal transport divergence between the predicted measure and a Gaussian-smoothed annotation measure. Soft correspondences replace hard assignment, and because the marginals are left free, total confidence mass survives as count supervision. A second training-only signal is drawn from the image itself predicted instances are rendered back to pixels through anisotropic Gaussian footprints, a shared blur kernel, pixel integration, and a capacity limited background, then compared with the observation under a Poisson–Gaussian noise model. Scene priors enter through a clipped log-space gate that caps how far context may suppress appearance evidence, a bound we audit by reporting recall on atypical placements alongside aggregate gains. The method reaches 31.7 AP on the AI-TOD-v2 validation split, including 23.4 AP on 2–8-pixel objects, and 36.8, 37.1, and 31.2 AP on SODA-D, SODA-A, and AI-TOD-R, with 53.6 AP on tiny persons in TinyPerson. Ablations isolate the contribution of each component.
More by Jasur Sevinov
- Physico-Analytical Modeling of an Industrial Low-Temperature Natural Gas Separation Line with Three-Section Cold Recovery and Propane CoolingProcesses · 2026
- Development of innovative approach for manufacturing pure composite basalt-glass plastic protective belt and transport pipesTechnology audit and production reserves · 2026
- Construction of the Principle of Adaptive Position-Trajectory Control of Moving Objects (On the Example of a Multi-Link Industrial Robot Manipulator)Cybernetics and Information Technologies · 2026
- A pole placement-based multimode modal and fuzzy-intelligent control approach for the ore grinding processInternational Conference on Geotechnology, Mining, and Rational Use of Natural Resources · 2026
