# Revisiting Binary Local Image Description for Resource Limited Devices

arXiv: https://arxiv.org/abs/2108.08380 · HTML: https://mlsanity.com/paper/2108.08380/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 41 references that are in the map.
- First posted on arXiv 18 Aug 2021.

## Builds on (most cited first, 8 of 8)

- [ORB-SLAM3: An Accurate Open-Source Library for Visual, Visual-Inertial and Multi-Map SLAM](https://mlsanity.com/paper/2007.11898/) (2020)
- [HPatches: A benchmark and evaluation of handcrafted and learned local descriptors](https://mlsanity.com/paper/1704.05939/) (2017)
- [Working hard to know your neighbor's margins: Local descriptor learning loss](https://mlsanity.com/paper/1705.10872/) (2017)
- [PL-SLAM: a Stereo SLAM System through the Combination of Points and Line Segments](https://mlsanity.com/paper/1705.09479/) (2017)
- [Local Descriptors Optimized for Average Precision](https://mlsanity.com/paper/1804.05312/) (2018)
- [LATCH: Learned Arrangements of Three Patch Codes](https://mlsanity.com/paper/1501.03719/) (2015)
- [HBST: A Hamming Distance embedding Binary Search Tree for Visual Place Recognition](https://mlsanity.com/paper/1802.09261/) (2018)
- [Supervised Hashing Using Graph Cuts and Boosted Decision Trees](https://mlsanity.com/paper/1408.5574/) (2014)

## Built on by (most cited first, 3 of 3)

- [GlueStick: Robust Image Matching by Sticking Points and Lines Together](https://mlsanity.com/paper/2304.02008/) (2023)
- [LightGlueStick: a Fast and Robust Glue for Joint Point-Line Matching](https://mlsanity.com/paper/2510.16438/) (2025)
- [Needle In A Haystack, Fast: Benchmarking Image Perceptual Similarity Metrics At Scale](https://mlsanity.com/paper/2206.00282/) (2022)

## Abstract (arXiv)

The advent of a panoply of resource limited devices opens up new challenges in the design of computer vision algorithms with a clear compromise between accuracy and computational requirements. In this paper we present new binary image descriptors that emerge from the application of triplet ranking loss, hard negative mining and anchor swapping to traditional features based on pixel differences and image gradients. These descriptors, BAD (Box Average Difference) and HashSIFT, establish new operating points in the state-of-the-art's accuracy vs. resources trade-off curve. In our experiments we evaluate the accuracy, execution time and energy consumption of the proposed descriptors. We show that BAD bears the fastest descriptor implementation in the literature while HashSIFT approaches in accuracy that of the top deep learning-based descriptors, being computationally more efficient. We have made the source code public.