3D mesh watermarking is an effective way to protect the copyright of 3D mesh models by imperceptibly embedding an ownership message into their geometric or topological information. The embedded watermark is supposed to be accurately extracted from the watermarked meshes after suffering various malicious or accidental geometric distortions during transmission, such as adding noise and smoothing. However, existing works rely on the geometric information of mesh models for watermark synchronization and extraction, resulting in poor robustness to geometric distortions. To address this issue, we propose a novel robust mesh watermarking scheme that is completely independent of geometric information. A watermark synchronization scheme is designed based on the Fiedler vector of the Kirchhoff matrix, upon which a watermark encoding strategy is introduced. Moreover, leveraging the vertex valence distribution of closed triangular meshes, we develop two topological embedding techniques to embed the watermark imperceptibly. Extensive experiments show that the proposed method achieves stronger robustness against geometric distortions, including similarity transformations, adding noise, smoothing, quantization, and even combination distortions, while maintaining the watermarked mesh quality compared with state-of-the-art methods.

Analog dynamical accelerators (DXs) are a growing sub-field in computer architecture research, offering order-of-magnitude gains in power efficiency and latency over traditional digital methods in several machine learning, optimization, and sampling tasks. However, limited-capacity accelerators require hybrid analog/digital algorithms to solve real-world problems, commonly using large-neighborhood local search (LNLS) frameworks. Unlike fully digital algorithms, hybrid LNLS has no non-asymptotic convergence guarantees and no principled hyperparameter selection schemes, particularly limiting cross-device training and inference. In this work, we provide non-asymptotic convergence guarantees for hybrid LNLS by reducing to block Langevin Diffusion (BLD) algorithms. Adapting tools from classical sampling theory, we prove exponential KL-divergence convergence for randomized and cyclic block selection strategies using ideal DXs. With finite device variation, we provide explicit bounds on the 2-Wasserstein bias in terms of step duration, noise strength, and function parameters. Our BLD model provides a key link between established theory and novel computing platforms, and our theoretical results provide a closed-form expression linking device variation, algorithm hyperparameters, and performance.

Nowadays, visual object detection (VOD) is widely used in many AI applications, such as autonomous driving, intelligent robotics, and smart surveillance. As an essential postprocessing step in VOD, nonmaximum suppression (NMS) is employed to generate bounding boxes as detection results. However, NMS is difficult to parallelize and computationally intensive, resulting in high processing latency and energy consumption. To address this issue, this brief proposes an energy-efficient block-based NMS engine that incorporates both algorithm- and hardware-level design techniques to improve processing speed and energy efficiency. These techniques include a block-NMS scheme, an adaptive hybrid sorting architecture (AHSA), and a reconfigurable pipeline-based block-NMS computation architecture. The proposed engine is implemented in 28-nm CMOS technology. Compared with the state-of-the-art designs, it achieves the highest performance (237.97 GOPS) and energy efficiency (8.71 TOPS/W), while delivering results fully equivalent to those of the original NMS.