Recent progress in hand gesture recognition has introduced several natural and intuitive approaches to drone control. However, effectively maneuvering drones in complex environments remains challenging. Drone movements are governed by four independent factors: roll, yaw, pitch, and throttle. Each factor includes three distinct behaviors—increase, decrease, and neutral—necessitating hand gesture vocabularies capable of expressing at least 81 combinations for comprehensive drone control in diverse scenarios. In this paper, we introduce a new set of hand gestures for precise drone control, leveraging an RGB camera sensor. These gestures are categorized into motion-based and posture-based types for efficient management. Then, we develop a lightweight hand gesture recognition algorithm capable of real-time operation on even edge devices, ensuring accurate and timely recognition. Subsequently, we integrate hand gesture recognition into a drone simulator to execute 81 commands for drone flight. Overall, the proposed hand gestures and recognition system offer natural control for complex drone maneuvers.
Recovering three-dimensional (3D) shape of an object from two-dimensional (2D) information is one of the major domains of computer vision applications. Shape from Focus (SFF) is a passive optical technique that reconstructs 3D shape of an object using 2D images with different focus settings. When a 2D image sequence is obtained with constant step size in SFF, mechanical vibrations, referred as jitter noise, occur in each step. Since the jitter noise changes the focus values of 2D images, it causes erroneous recovery of 3D shape. In this paper, a new filtering method for estimating optimal image positions is proposed. First, jitter noise is modeled as Gaussian or speckle function, secondly, the focus curves acquired by one of the focus measure operators are modeled as a quadratic function for application of the filter. Finally, Kalman filter as the proposed method is designed and applied for removing jitter noise. The proposed method is experimented by using image sequences of synthetic and real objects. The performance is evaluated through various metrics to show the effectiveness of the proposed method in terms of reconstruction accuracy and computational complexity. Root Mean Square Error (RMSE), correlation, Peak Signal-to-Noise Ratio (PSNR), and computational time of the proposed method are improved on average by about 48%, 11%, 15%, and 5691%, respectively, compared with conventional filtering methods.
5 May 2026 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Recent studies have shown that 2D convolution and self-attention exhibit distinct spectral behaviors, and optimizing their spectral properties can enhance vision model performance. However, theoretical analyses remain limited in explaining why 2D convolution is more effective in high-pass filtering than self-attention and why larger kernels favor shape bias, akin to self-attention. In this paper, we employ graph spectral analysis to theoretically simulate and compare the frequency responses of 2D convolution and self-attention within a unified framework. Our results corroborate previous empirical findings and reveal that node connectivity, modulated by window size, is a key factor in shaping spectral functions. Leveraging this insight, we introduce a spectral-adaptive modulation (SPAM) mixer, which processes visual features in a spectral-adaptive manner using multi-scale convolutional kernels and a spectral re-scaling mechanism to refine spectral components. Based on SPAM, we develop SPANetV2 as a novel vision backbone. Extensive experiments demonstrate that SPANetV2 outperforms state-of-the-art models across multiple vision tasks, including ImageNet-1K classification, COCO object detection, and ADE20K semantic segmentation.