awesome-visual-place-recognition

榜单

https://paperswithcode.com/task/visual-place-recognition

测评工具

https://github.com/gmberton/VPR-methods-evaluation

数据集

https://github.com/gmberton/VPR-datasets-downloader

https://github.com/amaralibey/gsv-cities

https://github.com/mapillary/mapillary_sls

ALTO: A Large-Scale Dataset for UAV Visual Place Recognition and Localization https://github.com/MetaSLAM/ALT

VPR-Bench： an open-source Visual Place Recognition evaluation framework with quantifiable viewpoint and illumination invariance. https://github.com/MubarizZaffar/VPR-Bench

Deep Visual Geo-localization Benchmark https://github.com/gmberton/deep-visual-geo-localization-benchmark

AmsterTime: A Visual Place Recognition Benchmark Dataset for Severe Domain Shift
https://github.com/seyrankhademi/AmsterTime

综述

General Place Recognition Survey: Towards the Real-world Autonomy Age https://github.com/MetaSLAM/GPRS

大模型方法

[cvpr2024][sota] BoQ: A Place is Worth a Bag of Learnable Queries [https://github.com/amaralibey/bag-of-queries]

[sota]EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition

【CVPR 2024】CricaVPR: Cross-image Correlation-aware Representation Learning for Visual Place Recognition https://github.com/Lu-Feng/CricaVPR

【ICLR 2024】[SelaVPR] Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition https://github.com/Lu-Feng/SelaVPR

ProGEO: Generating Prompts through Image-Text Contrastive Learning for Visual Geo-localization [https://github.com/chain-mao/progeo]

【SOTA】DINO-Mix: Enhancing Visual Place Recognition with Foundational Vision Model and Feature Mixing https://github.com/GaoShuang98/DINO-Mix

[sota] Optimal Transport Aggregation for Visual Place Recognition https://github.com/serizba/salad

AnyLoc: Towards Universal Visual Place Recognition
[https://github.com/AnyLoc/AnyLoc]

传统方法

【2013cvpr】all about vlad

Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences
https://github.com/ai4ce/NYU-VPR

cnn-based

netvlad

GEM Fine-tuning CNN Image Retrieval with No Human Annotation

ap loss ap-gem(DIR):Learning with Average Precision: Training Image Retrieval with a Listwise Loss https://github.com/naver/deep-image-retrieval

MultiRes-NetVLAD: Augmenting Place Recognition Training with Low-Resolution Imagery https://github.com/Ahmedest61/MultiRes-NetVLAD

Correlation Verification for Image Retrieval https://github.com/sungonce/CVNet

transformer based

LEARNING SUPER-FEATURES FOR IMAGE RETRIEVAL https://github.com/Vincentqyw/fire

TransVPR: Transformer-based place recognition with multi-level attention aggregation https://github.com/RuotongWANG/TransVPR-model-implementation

TransVLAD: Multi-Scale Attention-Based Global Descriptors for Visual Geo-Localization https://github.com/wacv-23/TVLAD

【sota wacv2023】MixVPR: Feature Mixing for Visual Place Recognition https://github.com/amaralibey/MixVPR

结合语义

【ICIAP 2021】Learning Semantics for Visual Place Recognition through Multi-Scale Attention https://github.com/valeriopaolicelli/SegVPR

StructVPR: Distill Structural Knowledge with Weighting Samples for Visual Place Recognition

A Novel Image Descriptor with Aggregated Semantic Skeleton Representation for Long-term Visual Place Recognition

轻量化

MobileNetVLAD [CoRL 2018]Leveraging Deep Visual Descriptors for Hierarchical Efficient Localization https://github.com/ethz-asl/hierarchical_loc

hfnet From Coarse to Fine: Robust Hierarchical Localization at Large Scale https://github.com/ethz-asl/hfnet

【2022 iros】LSDNet: A Lightweight Self-Attentional Distillation Network for Visual Place Recognition

局部特征结合

delf：Large-Scale Image Retrieval with Attentive Deep Local Features https://github.com/nashory/DeLF-pytorch

DELG Unifying Deep Local and Global Features for Image Search https://github.com/feymanpriv/DELG https://github.com/ditwoo/pytorch-delg-example

DOLG: Single-Stage Image Retrieval with Deep Orthogonal Fusion of Local and Global Features (ICCV 2021)
https://github.com/dongkyuk/DOLG-pytorch https://github.com/feymanpriv/DOLG https://github.com/feymanpriv/DOLG-paddle https://github.com/tanzeyy/DOLG-instance-retrieval

DALG: Deep Attentive Local and Global Modeling for Image Retrieval

loss与采样

Global Proxy-based Hard Mining for Visual Place Recognition（BMVC22） https://github.com/amaralibey/GPM

排序

[CVPR IMW 2023]Are Local Features All You Need for Cross-Domain Visual Place Recognition?
https://github.com/gbarbarani/re-ranking-for-VPR

Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition https://github.com/QVPR/Patch-NetVLAD

R2Former: Unified retrieval and reranking Transformer for Place Recognition. https://github.com/Jeff-Zilence/R2Former

【wacv2023】ETR: An Efficient Transformer for Re-Ranking in Visual Place Recognition https://github.com/HeartbreakSurvivor/ETR

Instance-level Image Retrieval using Reranking Transformers https://github.com/uvavision/RerankingTransformer

iccv2023 Global Features are All You Need for Image Retrieval and Reranking https://github.com/ShihaoShao-GH/SuperGlobal

标注方式

CamNet: Coarse-to-Fine Retrieval for Camera Re-Localization, ICCV 2019 https://github.com/dingmyu/CamNet

【sota cvpr2023 GCL】Data-efficient Large Scale Place Recognition with Graded Similarity Supervision https://github.com/marialeyvallina/generalized_contrastive_loss

【CVPR 2022 】Rethinking Visual Geo-localization for Large-Scale Applications https://github.com/gmberton/CosPlace

序列匹配

Fast and Memory Efficient Graph Optimization via ICM for Visual Place Recognition

【2012icra】SeqSLAM: Visual Route-Based Navigation for Sunny Summer Days and Stormy Winter Nights

https://github.com/tmadl/pySeqSLAM https://github.com/MetaSLAM/MRLoc https://github.com/subokita/OpenSeqSLAM https://github.com/siam1251/Fast-SeqSLAM

(2019 seqslam加速)MRS-VPR: a multi-resolution sampling based global visual place recognition method

SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition
SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition
https://github.com/oravus/seqNet

【CoRL 2021 Oral】SeqMatchNet: Contrastive Learning with Sequence Matching for Place Recognition and Relocalization
https://github.com/oravus/SeqMatchNet

SeqVLAD Learning Sequential Descriptors for Sequence-based Visual Place Recognition
https://github.com/vandal-vpr/vg-transformers

Lazy data association for image sequence matching under substantial appearance changes.
https://github.com/PRBonn/online_place_recognition

ICRA 2020 paper: Fast, Compact and Highly Scalable Visual Place Recognition through Sequence-based Matching of Overloaded Representations
https://github.com/youjiangxu/seqvlad-pytorch

Tracking‐DOSeqSLAM: A dynamic sequence‐based visual place recognition paradig
https://github.com/ktsintotas/tracking-DOSeqSLAM

DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition https://github.com/mchancan/deepseqslam

【2018TIP】Sequential Video VLAD: Training the Aggregation Locally and Temporally https://github.com/oravus/CoarseHash

Using Image Sequences for Long-Term Visual Localization https://github.com/rulllars/SequentialVisualLocalization

SeqOT: A Spatial-Temporal Transformer Network for Place Recognition Using Sequential LiDAR Data. https://github.com/BIT-MJY/SeqOT

STA-VPR: Spatio-temporal Alignment for Visual Place Recognition
https://github.com/Lu-Feng/STA-VPR

iccv23 Learning Sequence Descriptor based on Spatio-Temporal Attention for Visual Place Recognition

视角鲁棒性

EigenPlaces: Training Viewpoint Robust Models for Visual Place Recognition
https: //github.com/gmberton/EigenPlaces
while results with any other baseline can be computed with the codebase at https://github.com/gmberton/auto_VPR

全景

PanoVPR: Towards Unified Perspective-to-Equirectangular Visual Place Recognition via Sliding Windows across the Panoramic View
https://github.com/zafirshi/PanoVPR

交叉视角

TransGeo: Transformer Is All You Need for Cross-view Image Geo-localization
https://github.com/Jeff-Zilence/TransGeo2022

CVLNet: Cross-View Semantic Correspondence Learning for Video-based Camera Localization

域自适应

AdAGeo: Adaptive-Attentive Geolocalization from few queries: a hybrid approach https://github.com/valeriopaolicelli/adageo-WACV2021

city-wide VPR

Divide&Classify: Fine-Grained Classification for City-Wide Visual Place Recognition
https://github.com/ga1i13o/Divide-and-Classify

world-wide VPR

Where in the World is this Image? Transformer-based Geo-localization in the Wild
https://github.com/ShramanPramanick/Transformer_Based_Geo-localization

脉冲神经网络

VPRTempo: A Fast Temporally Encoded Spiking Neural Network for Visual Place Recognition