Semi-Supervised Domain Adaptation for Holistic Counting under Label Gap.

Abstract

This paper proposes a novel approach for semi-supervised domain adaptation for holistic regression tasks, where a DNN predicts a continuous value y∈R given an input image x . The current literature generally lacks specific domain adaptation approaches for this task, as most of them mostly focus on classification. In the context of holistic regression, most of the real-world datasets not only exhibit a covariate (or domain) shift, but also a label gap-the target dataset may contain labels not included in the source dataset (and vice versa). We propose an approach tackling both covariate and label gap in a unified training framework. Specifically, a Generative Adversarial Network (GAN) is used to reduce covariate shift, and label gap is mitigated via label normalisation. To avoid overfitting, we propose a stopping criterion that simultaneously takes advantage of the Maximum Mean Discrepancy and the GAN Global Optimality condition. To restore the original label range-that was previously normalised-a handful of annotated images from the target domain are used. Our experimental results, run on 3 different datasets, demonstrate that our approach drastically outperforms the state-of-the-art across the board. Specifically, for the cell counting problem, the mean squared error (MSE) is reduced from 759 to 5.62; in the case of the pedestrian dataset, our approach lowered the MSE from 131 to 1.47. For the last experimental setup, we borrowed a task from plant biology, i.e., counting the number of leaves in a plant, and we ran two series of experiments, showing the MSE is reduced from 2.36 to 0.88 (intra-species), and from 1.48 to 0.6 (inter-species).

Plant phenotyping relevance

植物の葉数という形態形質を画像から推定する計算手法を開発し、種内・種間で性能評価している。植物課題は複数実験の一部だが、ドメイン適応と葉数カウントの技術評価が明示されており、単なるルーチン測定ではない。

Code and data availability

The paper's authors explicitly state their implementation code is publicly available on GitHub; this is the authors' analysis code for the leaf/cell/pedestrian counting domain adaptation experiments. The CVPPP (Zenodo) and other datasets are cited prior-work datasets, not paper-specific assets.