Object detection in remote sensing images relies on a large amount of labeled data for training. The growing new categories and class imbalance render exhaustive annotation nonscalable. Few-shot object detection (FSOD) tackles this issue by meta-learning on seen base classes and then fine-tuning on novel classes with few labeled samples. However, the object's scale and orientation variations are particularly large in remote sensing images, thus posing challenges to existing few-shot object detection methods. To tackle these challenges, we first propose to integrate a feature pyramid network and use prototype features to highlight query features to improve upon existing FSOD methods. We refer to the modified FSOD as a Strong Baseline which is demonstrated to perform significantly better than the original baselines. To improve the robustness of orientation variation, we further propose a transformation-invariant network (TINet) to allow the network to be invariant to geometric transformations. Extensive experiments on three widely used remote sensing object detection datasets, i.e., NWPU VHR-10.v2, DIOR, and HRRSD demonstrated the effectiveness of the proposed method. Finally, we reproduced multiple FSOD methods for remote sensing images to create an extensive benchmark for follow-up works.
We demonstrate how programmable shape evolution and deformation can be induced in plant-based natural materials through standard digital printing technologies. With nonallergenic pollen paper as the substrate material, we show how specific geometrical features and architectures can be custom designed through digital printing of patterns to modulate hygrophobicity, geometry, and complex shapes. These autonomously hygromorphing configurations can be “frozen” by postprocessing coatings to meet the needs of a wide spectrum of uses and applications. Through computational simulations involving the finite element method and accompanying experiments, we develop quantitative insights and a general framework for creating complex shapes in eco-friendly natural materials with potential sustainable applications for scalable manufacturing.
There is tremendous interest in developing 3D scaffolds from natural materials for a wide range of healthcare, energy, photonic, and environmental science applications. To date, most natural materials that are used to make 3D scaffolds consist of fibril structures; however, it would be advantageous to explore the development of scaffolds from natural materials with distinct supramolecular structures. Herein, the fabrication of a mechanically responsive pollen sponge that exhibits tunable 3D scaffold properties and is useful for oil remediation applications is reported. By using pollen‐based microgel particles as colloidal building blocks, the sponge fabrication process is optimized by tuning the processing conditions during freeze‐drying and thermal annealing steps. Stearic acid functionalization transforms the pollen sponge into a hydrophobic scaffold that can readily and repeatedly absorb oil and other organic solvents from contaminated water sources, with similar performance levels to commercial, synthetic polymer‐based absorbents and an improved environmental footprint.
The development of multifunctional 3D printing materials from sustainable natural resources is a high priority in additive manufacturing. Using an eco-friendly method to transform hard pollen grains into stimulus-responsive microgel particles, we engineered a pollen-derived microgel suspension that can serve as a functional reinforcement for composite hydrogel inks and as a supporting matrix for versatile freeform 3D printing systems. The pollen microgel particles enabled the printing of composite inks and improved the mechanical and physiological stabilities of alginate and hyaluronic acid hydrogel scaffolds for 3D cell culture applications. Moreover, the particles endowed the inks with stimulus-responsive controlled release properties. The suitability of the pollen microgel suspension as a supporting matrix for freeform 3D printing of alginate and silicone rubber inks was demonstrated and optimized by tuning the rheological properties of the microgel. Compared with other classes of natural materials, pollen grains have several compelling features, including natural abundance, renewability, affordability, processing ease, monodispersity, and tunable rheological features, which make them attractive candidates to engineer advanced materials for 3D printing applications.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.