Illumination Invariant Imaging : Applications in Robust Vision-based Localisation , Mapping and Classification for Autonomous Vehicles

By William P. Maddern, Alexander D. Stewart, Colin McManus, Ben Upcroft, Winston Churchill, Paul M. Newman. 2014

146 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A Survey of Autonomous Driving: Common Practices and Emerging Technologies (2019, positive): “Pre-processed, illumination invariant images have applied to ADS [149] and were shown to improve localization, mapping and scene classification capabilities over long periods of time.”
It's not easy seeing green: Lighting-resistant stereo Visual Teach & Repeat using color-constant images (2015, unknown): “This path spans harsh unstructured environments such as rocks and sand, windy grasslands, and strongly shadowed wooded areas.”
Visual Place Recognition: A Survey (2016, unknown): “assumptions about black-body illumination are violated [208].”
24/7 Place Recognition by View Synthesis (2018, unknown): “[51] generate illumination invariant images by conversion from 3-channel (RGB) to a single channel image using peaks of spectral responses of the sensor.”
Place Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free (2015, methodology): “Place Recognition: Methods that address the place recognition problem span from matching sequences of images [27, 17, 40, 33, 29], transforming images to become invariant against common scene changes such as shadows [6, 43, 25, 24, 21], learning how environments change over time and predicting these”
Deep learning features at scale for visual place recognition (2017, background): “Visual place recognition over perceptually-changing environments generally falls into two categories: utilizing feature representations that are robust to perceptual changes [11, 14-16] or learning and predicting appearance changes [17-20].”
1 year, 1000 km: The Oxford RobotCar dataset (2017, methodology): “For example, methods that depend on photometric error as in Engel et al. (2014) or precise colorimetry as in Maddern et al. (2014) are strongly affected by image compression and Bayer demosaicing; by logging raw packets directly from the camera we minimize the restrictions on future research.”