# Learning and Crafting for the Wide Multiple Baseline Stereo

arXiv: https://arxiv.org/abs/2112.12027 · HTML: https://mlsanity.com/paper/2112.12027/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 303 references that are in the map.
- First posted on arXiv 22 Dec 2021.

## Builds on (most cited first, 9 of 9)

- [Billion-scale similarity search with GPUs](https://mlsanity.com/paper/1702.08734/) (2017)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [In Defense of the Triplet Loss for Person Re-Identification](https://mlsanity.com/paper/1703.07737/) (2017)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [Night-to-Day Image Translation for Retrieval-based Localization](https://mlsanity.com/paper/1809.09767/) (2018)
- [Toward Geometric Deep SLAM](https://mlsanity.com/paper/1707.07410/) (2017)
- [Making Affine Correspondences Work in Camera Geometry Computation](https://mlsanity.com/paper/2007.10032/) (2020)
- [Minimal Cases for Computing the Generalized Relative Pose using Affine Correspondences](https://mlsanity.com/paper/2007.10700/) (2020)
- [In the Saddle: Chasing Fast and Repeatable Features](https://mlsanity.com/paper/1608.06800/) (2016)

## Built on by (most cited first, 1 of 1)

- [OpenGlue: Open Source Graph Neural Net Based Pipeline for Image Matching](https://mlsanity.com/paper/2204.08870/) (2022)

## Abstract (arXiv)

This thesis introduces the wide multiple baseline stereo (WxBS) problem. WxBS, a generalization of the standard wide baseline stereo problem, considers the matching of images that simultaneously differ in more than one image acquisition factor such as viewpoint, illumination, sensor type, or where object appearance changes significantly, e.g., over time. A new dataset with the ground truth, evaluation metric and baselines has been introduced. The thesis presents the following improvements of the WxBS pipeline. (i) A loss function, called HardNeg, for learning a local image descriptor that relies on hard negative mining within a mini-batch and on the maximization of the distance between the closest positive and the closest negative patches. (ii) The descriptor trained with the HardNeg loss, called HardNet, is compact and shows state-of-the-art performance in standard matching, patch verification and retrieval benchmarks. (iii) A method for learning the affine shape, orientation, and potentially other parameters related to geometric and appearance properties of local features. (iv) A tentative correspondences generation strategy which generalizes the standard first to second closest distance ratio is presented. The selection strategy, which shows performance superior to the standard method, is applicable to either hard-engineered descriptors like SIFT, LIOP, and MROGH or deeply learned like HardNet. (v) A feedback loop is introduced for the two-view matching problem, resulting in MODS -- matching with on-demand view synthesis -- algorithm. MODS is an algorithm that handles a viewing angle difference even larger than the previous state-of-the-art ASIFT algorithm, without a significant increase of computational cost over "standard" wide and narrow baseline approaches. Last, but not least, a comprehensive benchmark for local features and robust estimation algorithms is introduced.