D$^{2}$-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation

• URL: http://arxiv.org/abs/2511.12528v1
• Date: Sun, 16 Nov 2025 09:47:45 GMT
• Title: D$^{2}$-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation
• Authors: Zheyuan Zhang, Jiwei Zhang, Boyu Zhou, Linzhimeng Duan, Hong Chen,
• Abstract summary: Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged database.<n>DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance.<n>We propose $D2$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models.
• Score: 21.709098547489692
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2's exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2% and FLOPs by about 62.6% (compared to CricaVPR).Code is available at https://github.com/tony19980810/D2VPR.

Related papers

• Beyond Weight Adaptation: Feature-Space Domain Injection for Cross-Modal Ship Re-Identification [3.6907522136316975]
  Cross-Modality Ship Re-Identification (CMS Re-ID) is critical for achieving all-day and all-weather maritime target tracking.<n>We explore the potential of Vision Foundation Models (VFMs) in bridging modality gaps.<n>We propose a novel PEFT strategy termed Domain Representation Injection (DRI)
  arXiv  Detail & Related papers  (2025-12-24T02:30:23Z)
• MobileGeo: Exploring Hierarchical Knowledge Distillation for Resource-Efficient Cross-view Drone Geo-Localization [47.16612614191333]
  Cross-view geo-localization enables drone localization by matching aerial images to geo-tagged satellite databases.<n>MobileGeo is a mobile-friendly framework designed for efficient on-device CVGL.<n>MobileGeo runs at 251.5 FPS on an NVIDIA AGX Orin edge device, demonstrating its practical viability for real-time on-device drone geo-localization.
  arXiv  Detail & Related papers  (2025-10-26T08:47:20Z)
• SimpleGVR: A Simple Baseline for Latent-Cascaded Video Super-Resolution [46.311223206965934]
  We study key design principles for latter cascaded video super-resolution models, which are underexplored currently.<n>First, we propose two strategies to generate training pairs that better mimic the output characteristics of the base model, ensuring alignment between the VSR model and its upstream generator.<n>Second, we provide critical insights into VSR model behavior through systematic analysis of (1) timestep sampling strategies, (2) noise augmentation effects on low-resolution (LR) inputs.
  arXiv  Detail & Related papers  (2025-06-24T17:57:26Z)
• KARE-RAG: Knowledge-Aware Refinement and Enhancement for RAG [63.82127103851471]
  Retrieval-Augmented Generation (RAG) enables large language models to access broader knowledge sources.<n>We demonstrate that enhancing generative models' capacity to process noisy content is equally critical for robust performance.<n>We present KARE-RAG, which improves knowledge utilization through three key innovations.
  arXiv  Detail & Related papers  (2025-06-03T06:31:17Z)
• Reinforced Model Merging [53.84354455400038]
  We present an innovative framework termed Reinforced Model Merging (RMM), which encompasses an environment and agent tailored for merging tasks.<n>By utilizing data subsets during the evaluation process, we addressed the bottleneck in the reward feedback phase, thereby accelerating RMM by up to 100 times.
  arXiv  Detail & Related papers  (2025-03-27T08:52:41Z)
• BEVDiffLoc: End-to-End LiDAR Global Localization in BEV View based on Diffusion Model [8.720833232645155]
  Bird's-Eye-View (BEV) image is one of the most widely adopted data representations in autonomous driving.<n>We propose BEVDiffLoc, a novel framework that formulates LiDAR localization as a conditional generation of poses.
  arXiv  Detail & Related papers  (2025-03-14T13:17:43Z)
• SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition [91.98099115144511]
  Recent studies show that the visual place recognition (VPR) method using pre-trained visual foundation models can achieve promising performance.<n>We propose a novel method to realize seamless adaptation of foundation models to VPR.<n>In pursuit of higher efficiency and better performance, we propose an extension of the SelaVPR, called SelaVPR++.
  arXiv  Detail & Related papers  (2025-02-23T15:01:09Z)
• ALoRE: Efficient Visual Adaptation via Aggregating Low Rank Experts [71.91042186338163]
  ALoRE is a novel PETL method that reuses the hypercomplex parameterized space constructed by Kronecker product to Aggregate Low Rank Experts.<n>Thanks to the artful design, ALoRE maintains negligible extra parameters and can be effortlessly merged into the frozen backbone.
  arXiv  Detail & Related papers  (2024-12-11T12:31:30Z)
• Exploiting Distribution Constraints for Scalable and Efficient Image Retrieval [1.6874375111244329]
  State-of-the-art image retrieval systems train specific neural networks for each dataset.<n>Off-the-shelf foundation models fall short in achieving performance comparable to dataset-specific models.<n>We introduce Autoencoders with Strong Variance Constraints (AE-SVC), which significantly improves the performance of foundation models.
  arXiv  Detail & Related papers  (2024-10-09T16:05:16Z)
• EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition [6.996304653818122]
  We present an effective approach to harness the potential of a foundation model for Visual Place Recognition.<n>We show that features extracted from self-attention layers can act as a powerful re-ranker for VPR, even in a zero-shot setting.<n>Our method also demonstrates exceptional robustness and generalization, setting new state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-05-28T11:24:41Z)
• When Parameter-efficient Tuning Meets General-purpose Vision-language Models [65.19127815275307]
  PETAL revolutionizes the training process by requiring only 0.5% of the total parameters, achieved through a unique mode approximation technique. Our experiments reveal that PETAL not only outperforms current state-of-the-art methods in most scenarios but also surpasses full fine-tuning models in effectiveness.
  arXiv  Detail & Related papers  (2023-12-16T17:13:08Z)
• EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval [83.79667141681418]
  Large neural models (such as Transformers) achieve state-of-the-art performance for information retrieval (IR) We propose a novel distillation approach that leverages the relative geometry among queries and documents learned by the large teacher model. We show that our approach successfully distills from both dual-encoder (DE) and cross-encoder (CE) teacher models to 1/10th size asymmetric students that can retain 95-97% of the teacher performance.
  arXiv  Detail & Related papers  (2023-01-27T22:04:37Z)
• On Exploring Pose Estimation as an Auxiliary Learning Task for Visible-Infrared Person Re-identification [66.58450185833479]
  In this paper, we exploit Pose Estimation as an auxiliary learning task to assist the VI-ReID task in an end-to-end framework. By jointly training these two tasks in a mutually beneficial manner, our model learns higher quality modality-shared and ID-related features. Experimental results on two benchmark VI-ReID datasets show that the proposed method consistently improves state-of-the-art methods by significant margins.
  arXiv  Detail & Related papers  (2022-01-11T09:44:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.