# SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition

arXiv: https://arxiv.org/abs/2106.11481 · HTML: https://mlsanity.com/paper/2106.11481/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 34 references that are in the map.
- First posted on arXiv 22 Jun 2021.

## Builds on (most cited first, 21 of 21)

- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [D2-Net: A Trainable CNN for Joint Detection and Description of Local Features](https://mlsanity.com/paper/1905.03561/) (2019)
- [PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition](https://mlsanity.com/paper/1804.03492/) (2018)
- [Learning with Average Precision: Training Image Retrieval with a Listwise Loss](https://mlsanity.com/paper/1906.07589/) (2019)
- [Deep Learning Features at Scale for Visual Place Recognition](https://mlsanity.com/paper/1701.05105/) (2017)
- [Where is your place, Visual Place Recognition?](https://mlsanity.com/paper/2103.06443/) (2021)
- [SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition](https://mlsanity.com/paper/2102.11603/) (2021)
- [Semantics for Robotic Mapping, Perception and Interaction: A Survey](https://mlsanity.com/paper/2101.00443/) (2021)
- [DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DoF Relocalization](https://mlsanity.com/paper/2007.09217/) (2020)
- [Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Pooling](https://mlsanity.com/paper/2011.14497/) (2020)
- [FutureMapping: The Computational Structure of Spatial AI Systems](https://mlsanity.com/paper/1803.11288/) (2018)
- [LoST? Appearance-Invariant Place Recognition for Opposite Viewpoints using Visual Semantics](https://mlsanity.com/paper/1804.05526/) (2018)
- [Leveraging Deep Visual Descriptors for Hierarchical Efficient Localization](https://mlsanity.com/paper/1809.01019/) (2018)
- [Augmenting Visual Place Recognition with Structural Cues](https://mlsanity.com/paper/2003.00278/) (2020)
- [Delta Descriptors: Change-Based Place Representation for Robust Visual Localization](https://mlsanity.com/paper/2006.05700/) (2020)
- [Efficient 3D Point Cloud Feature Learning for Large-Scale Place Recognition](https://mlsanity.com/paper/2101.02374/) (2021)
- [Condition-Invariant Multi-View Place Recognition](https://mlsanity.com/paper/1902.09516/) (2019)
- [Look No Deeper: Recognizing Places from Opposing Viewpoints under Varying Scene Appearance using Single-View Depth Estimation](https://mlsanity.com/paper/1902.07381/) (2019)
- [DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition](https://mlsanity.com/paper/2011.08518/) (2020)

## Abstract (arXiv)

Place Recognition is a crucial capability for mobile robot localization and navigation. Image-based or Visual Place Recognition (VPR) is a challenging problem as scene appearance and camera viewpoint can change significantly when places are revisited. Recent VPR methods based on ``sequential representations'' have shown promising results as compared to traditional sequence score aggregation or single image based techniques. In parallel to these endeavors, 3D point clouds based place recognition is also being explored following the advances in deep learning based point cloud processing. However, a key question remains: is an explicit 3D structure based place representation always superior to an implicit ``spatial'' representation based on sequence of RGB images which can inherently learn scene structure. In this extended abstract, we attempt to compare these two types of methods by considering a similar ``metric span'' to represent places. We compare a 3D point cloud based method (PointNetVLAD) with image sequence based methods (SeqNet and others) and showcase that image sequence based techniques approach, and can even surpass, the performance achieved by point cloud based methods for a given metric span. These performance variations can be attributed to differences in data richness of input sensors as well as data accumulation strategies for a mobile robot. While a perfect apple-to-apple comparison may not be feasible for these two different modalities, the presented comparison takes a step in the direction of answering deeper questions regarding spatial representations, relevant to several applications like Autonomous Driving and Augmented/Virtual Reality. Source code available publicly https://github.com/oravus/seqNet.