CLIDD: Cross-Layer Independent Deformable Description for Efficient and Discriminative Local Feature Representation

TL;DR

CLIDD introduces a novel cross-layer deformable descriptor that enhances local feature discriminativeness and efficiency, enabling real-time spatial intelligence applications with minimal computational resources.

Contribution

The paper presents CLIDD, a new method combining independent feature hierarchies and hardware-aware optimization for scalable, high-performance local feature descriptors.

Findings

Ultra-compact model matches SuperPoint accuracy with 0.004M parameters.

High-performance variant exceeds 200 FPS on edge devices.

Achieves superior matching accuracy and efficiency compared to state-of-the-art methods.

Abstract

Robust local feature representations are essential for spatial intelligence tasks such as robot navigation and augmented reality. Establishing reliable correspondences requires descriptors that provide both high discriminative power and computational efficiency. To address this, we introduce Cross-Layer Independent Deformable Description (CLIDD), a method that achieves superior distinctiveness by sampling directly from independent feature hierarchies. This approach utilizes learnable offsets to capture fine-grained structural details across scales while bypassing the computational burden of unified dense representations. To ensure real-time performance, we implement a hardware-aware kernel fusion strategy that maximizes inference throughput. Furthermore, we develop a scalable framework that integrates lightweight architectures with a training protocol leveraging both metric learning and knowledge distillation. This scheme generates a wide spectrum of model variants optimized for diverse deployment constraints. Extensive evaluations demonstrate that our approach achieves superior matching accuracy and exceptional computational efficiency simultaneously. Specifically, the ultra-compact variant matches the precision of SuperPoint while utilizing only 0.004M parameters, achieving a 99.7% reduction in model size. Furthermore, our high-performance configuration outperforms all current state-of-the-art methods, including high-capacity DINOv2-based frameworks, while exceeding 200 FPS on edge devices. These results demonstrate that CLIDD delivers high-precision local feature matching with minimal computational overhead, providing a robust and scalable solution for real-time spatial intelligence tasks.