Learning Condition Invariant Features For Retrieval-based Localization From 1M Images

Abstract

Image features for retrieval-based localization must be invariant to dynamic objects (e.g. cars) as well as seasonal and daytime changes. Such invariances are, up to some extent, learnable with existing methods using triplet-like losses, given a large number of diverse training images. However, due to the high algorithmic training complexity, there exists insufficient comparison between different loss functions on large datasets. In this paper, we train and evaluate several localization methods on three different benchmark datasets, including Oxford RobotCar with over one million images. This large scale evaluation yields valuable insights into the generalizability and performance of retrieval-based localization. Based on our findings, we develop a novel method for learning more accurate and better generalizing localization features. It consists of two main contributions: (i) a feature volume-based loss function, and (ii) hard positive and pairwise negative mining. On the challenging O

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keythoma2020learning

Related papers

• Retrieval-based Localization Based On Domain-invariant Feature Learning Under Changing Environments (2019)9.92
• Domain-invariant Similarity Activation Map Contrastive Learning For Retrieval-based Long-term Visual Localization (2020)13.72
• Geometrically Mappable Image Features (2020)6.34
• Adversarial Training For Adverse Conditions: Robust Metric Localisation Using Appearance Transfer (2018)14.43
• Improving Nighttime Retrieval-based Localization (2018)0.00
• Deep Dense Local Feature Matching And Vehicle Removal For Indoor Visual Localization (2022)0.00
• Benchmarking Image Retrieval For Visual Localization (2020)17.78
• Investigating The Role Of Image Retrieval For Visual Localization -- An Exhaustive Benchmark (2022)16.58