InLoc: Indoor Visual Localization with Dense Matching and View Synthesis

By Hajime Taira, Masatoshi Okutomi, Torsten Sattler, Mircea Cimpoi, Marc Pollefeys, Josef Sivic, Tomás Pajdla, Akihiko Torii. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019

555 citations recorded by the source corpus.

We seek to predict the 6 degree-of-freedom (6DoF) pose of a query photograph with respect to a large indoor 3D map. The contributions of this work are three-fold. First, we develop a new large-scale visual localization method targeted for indoor spaces. The method proceeds along three steps: (i) efficient retrieval of candidate poses that scales to large-scale environments, (ii) pose estimation using dense matching rather than sparse local features to deal with weakly textured indoor scenes, and (iii) pose verification by virtual view synthesis that is robust to significant changes in viewpoint, scene layout, and occlusion. Second, we release a new dataset with reference 6DoF poses for large-scale indoor localization. Query photographs are captured by mobile phones at a different time than the reference 3D map, thus presenting a realistic indoor localization scenario. Third, we demonstrate that our method significantly outperforms current state-of-the-art indoor localization approaches on this new challenging data. Code and data are publicly available.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/6332673870ababfd414342723d5f220179573c25

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Unifying Deep Local and Global Features for Image Search (2020, methodology): “[55] used NetVLAD [1] to extract global features for candidate pose retrieval, followed by dense local feature matching using feature maps from the same network.”
Visual Camera Re-Localization From RGB and RGB-D Images Using DSAC (2020, methodology): “We compare to the following relative pose estimation approaches: AnchorNet [31], and retrieval-based InLoc [25].”
RoMa: Robust Dense Feature Matching (2023, unknown): “We evaluate RoMa on the InLoc [45] Visual Localization benchmark, using the HLoc [40] pipeline.”
Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition (2021, methodology): “Local Keypoint Descriptors: Local keypoint methods are often used to re-rank initial place match candidate lists produced by a global approach [23, 74, 61].”
LoFTR: Detector-Free Local Feature Matching with Transformers (2021, methodology): “It benchmarks on two datasets, the AachenDay-Night dataset [38, 54] concerning outdoor scenes and the InLoc [41] dataset concerning indoor scenes.”
LightGlue: Local Feature Matching at Light Speed (2023, methodology): “Failure cases on InLoc [70].”
ASpanFormer: Detector-Free Image Matching with Adaptive Span Transformer (2022, methodology): “Apart from evaluation on two-view pose estimation task, we further integrate our network into a visual localization pipeline, and use two popular datasets, InLoc [53] and Aachen Day-Night v1.”
A review of multimodal image matching: Methods and applications (2021, unknown): “[511] presented InLoc for indoor visual localization.”
Novel Visual Category Discovery with Dual Ranking Statistics and Mutual Knowledge Distillation (2021, background): “Part-level features have been more widely used in the domain of image retrieval with both classical [34, 5] and deep learning based [2, 37, 42, 45, 7] methods.”
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features (2019, methodology): “For this experiment, we integrated our features into two variants of the pipeline proposed in [58], using the code released by the authors.”
Understanding the Limitations of CNN-Based Absolute Camera Pose Regression (2019, methodology): “These matches are established by descriptor matching [21, 37, 39, 59, 63, 68, 69, 81] or by regressing 3D coordinates from pixel patches [8–10, 16, 23, 43, 46, 47, 65].”
Image Matching Across Wide Baselines: From Paper to Practice (2020, background): “It also performs poorly overall in our evaluation, despite reporting state-of-the-art results in other benchmarks (Mikolajczyk and Schmid 2004; Balntas et al. 2017; Sattler et al. 2012; Taira et al. 2019)— without the ratio test, the number of tentative matches might be too high for RANSAC or COLMAP”
Grounding Image Matching in 3D with MASt3R (2024, methodology): “We provide here additional qualitative results on the DTU [3], InLoc [84], Aachen Day-Night datasets [118] and the Map-free benchmark [5].”