Image-Conditioned Adaptive Parameter Tuning for Visual Odometry Frontends

• URL: http://arxiv.org/abs/2603.21785v1
• Date: Mon, 23 Mar 2026 10:26:25 GMT
• Title: Image-Conditioned Adaptive Parameter Tuning for Visual Odometry Frontends
• Abstract summary: Resource-constrained autonomous robots rely on sparse direct and semi-direct visual-(inertial)-odometry (VO) pipelines.<n>We propose the first image-conditioned reinforcement learning framework for online tuning of VO parameters.
• Score: 21.343239924094764
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Resource-constrained autonomous robots rely on sparse direct and semi-direct visual-(inertial)-odometry (VO) pipelines, as they provide a favorable tradeoff between accuracy, robustness, and computational cost. However, the performance of most systems depends critically on hand-tuned hyperparameters governing feature detection, tracking, and outlier rejection. These parameters are typically fixed during deployment, even though their optimal values vary with scene characteristics such as texture density, illumination, motion blur, and sensor noise, leading to brittle performance in real-world environments. We propose the first image-conditioned reinforcement learning framework for online tuning of VO frontend parameters, effectively embedding the expert into the system. Our key idea is to formulate the frontend configuration as a sequential decision-making problem and learn a policy that directly maps visual input to feature detection and tracking parameters. The policy uses a lightweight texture-aware CNN encoder and a privileged critic during training. Unlike prior RL-based approaches that rely solely on internal VO statistics, our method observes the image content and proactively adapts parameters before tracking degrades. Experiments on TartanAirV2 and TUM RGB-D show 3x longer feature tracks and 3x lower computational cost, despite training entirely in simulation.

Related papers

• A Physically-Grounded Attack and Adaptive Defense Framework for Real-World Low-Light Image Enhancement [11.903420482259257]
  Limited illumination often causes severe physical noise and detail degradation in images.<n>Existing Low-Light Image Enhancement (LLIE) methods treat the enhancement process as a blind black-box mapping.<n>We propose a novel LLIE approach, conceptually formulated as a physics-based attack and display-adaptive defense paradigm.
  arXiv  Detail & Related papers  (2026-03-15T09:44:54Z)
• Taming Vision Priors for Data Efficient mmWave Channel Modeling [17.020987773588907]
  Differentiable ray tracing offers a physics-grounded solution for modeling millimeter-wave (mmWave) propagation.<n>Differentiable ray tracing offers a physics-grounded solution but still facing deployment challenges due to its over-reliance on exhaustive channel measurements.<n>We present VisRFTwin, a scalable and data-efficient digital-twin framework that integrates vision-derived material priors with differentiable ray tracing.
  arXiv  Detail & Related papers  (2026-03-11T04:35:51Z)
• Depth Completion as Parameter-Efficient Test-Time Adaptation [66.72360181325877]
  CAPA is a parameter-efficient test-time optimization framework that adapts pre-trained 3D foundation models (FMs) for depth completion.<n>For videos, CAPA introduces sequence-level parameter sharing, jointly adapting all frames to exploit temporal correlations, improve robustness, and enforce multi-frame consistency.
  arXiv  Detail & Related papers  (2026-02-16T13:53:23Z)
• PPISP: Physically-Plausible Compensation and Control of Photometric Variations in Radiance Field Reconstruction [10.040461884848403]
  Physically-Plausible ISP correction module disentangles camera-intrinsic and capture-dependent effects.<n>A dedicated PPISP controller, trained on the input views, predicts ISP parameters for novel viewpoints.
  arXiv  Detail & Related papers  (2026-01-26T10:23:43Z)
• Analysis of Hyperparameter Optimization Effects on Lightweight Deep Models for Real-Time Image Classification [0.0]
  This study evaluates the accuracy and deployment feasibility of seven modern lightweight architectures: ConvNeXt-T, EfficientV2-S, MobileNetV3-L, MobileViT v2 (S/XS), RepVGG-A2, and TinyViT-21M.<n> tuning alone leads to a top-1 accuracy improvement of 1.5 to 3.5 percent over baselines, and select models deliver latency under 5 milliseconds and over 9,800 frames per second.
  arXiv  Detail & Related papers  (2025-07-31T07:47:30Z)
• Continual Adaptation: Environment-Conditional Parameter Generation for Object Detection in Dynamic Scenarios [54.58186816693791]
  environments constantly change over time and space, posing significant challenges for object detectors trained based on a closed-set assumption.<n>We propose a new mechanism, converting the fine-tuning process to a specific- parameter generation.<n>In particular, we first design a dual-path LoRA-based domain-aware adapter that disentangles features into domain-invariant and domain-specific components.
  arXiv  Detail & Related papers  (2025-06-30T17:14:12Z)
• Visual Fourier Prompt Tuning [63.66866445034855]
  We propose the Visual Fourier Prompt Tuning (VFPT) method as a general and effective solution for adapting large-scale transformer-based models. Our approach incorporates the Fast Fourier Transform into prompt embeddings and harmoniously considers both spatial and frequency domain information. Our results demonstrate that our approach outperforms current state-of-the-art baselines on two benchmarks.
  arXiv  Detail & Related papers  (2024-11-02T18:18:35Z)
• Exploring Dynamic Transformer for Efficient Object Tracking [58.120191254379854]
  We propose DyTrack, a dynamic transformer framework for efficient tracking.<n>DyTrack automatically learns to configure proper reasoning routes for various inputs, gaining better utilization of the available computational budget.<n>Experiments on multiple benchmarks demonstrate that DyTrack achieves promising speed-precision trade-offs with only a single model.
  arXiv  Detail & Related papers  (2024-03-26T12:31:58Z)
• Learning IMM Filter Parameters from Measurements using Gradient Descent [45.335821132209766]
  In intrinsic parameters of targets under track can be completely unobservable until the system is deployed. With state-of-the-art sensor systems growing more and more complex, the number of parameters naturally increases. In this paper, the parameters of an interacting multiple model (IMM) filter are optimized solely using measurements.
  arXiv  Detail & Related papers  (2023-07-13T08:35:40Z)
• Sensitivity-Aware Visual Parameter-Efficient Fine-Tuning [91.5113227694443]
  We propose a novel visual. sensuous-aware fine-Tuning (SPT) scheme. SPT allocates trainable parameters to task-specific important positions. Experiments on a wide range of downstream recognition tasks show that our SPT is complementary to the existing PEFT methods.
  arXiv  Detail & Related papers  (2023-03-15T12:34:24Z)
• Deep Learning based Virtual Point Tracking for Real-Time Target-less Dynamic Displacement Measurement in Railway Applications [0.0]
  We propose virtual point tracking for real-time target-less dynamic displacement measurement, incorporating deep learning techniques and domain knowledge. We demonstrate our approach for a railway application, where the lateral displacement of the wheel on the rail is measured during operation.
  arXiv  Detail & Related papers  (2021-01-17T16:19:47Z)
• Rethinking the Hyperparameters for Fine-tuning [78.15505286781293]
  Fine-tuning from pre-trained ImageNet models has become the de-facto standard for various computer vision tasks. Current practices for fine-tuning typically involve selecting an ad-hoc choice of hyper parameters. This paper re-examines several common practices of setting hyper parameters for fine-tuning.
  arXiv  Detail & Related papers  (2020-02-19T18:59:52Z)

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.