Pith. sign in

REVIEW 3 major objections 6 minor 46 references

Reproducible Evaluation of Camera Auto-Exposure Methods in the Field: Platform, Benchmark and Lessons Learned

T0 review · 3 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Camera auto-exposure methods can be benchmarked reproducibly offline, and a classical brightness-targeting controller still comes out on top.

desk verdict A valuable multi-exposure dataset and honest static validation, but the AE benchmark ranking needs uncertainty bounds and dynamic validation before the headline claim is taken at face value. read the letter →

arxiv 2506.18844 v1 pith:ZHP3IFMP submitted 2025-06-19 cs.RO cs.CV

Olivier Gamache , Jean-Michel Fortin , Matěj Boxan , François Pomerleau , Philippe Giguère This is my paper · ORCID
classification cs.ROcs.CV
keywords auto-exposureexposureemulationcameraresponsefunctionvisualSLAMbenchmarkingmulti-exposuredatasetBorealHDRfieldrobotics
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Auto-exposure (AE) methods adjust camera exposure while the robot runs, which is why they have traditionally been tested in the field with live hardware and cannot be replayed. The paper's proposal is to remove the hardware from the loop: from a burst of bracket exposures of the same scene, a camera response function can remap pixels to any intermediate exposure time offline, producing photo-realistic images after the fact. Static validation against 1000 ground-truth images per scene gives a median error of 0.21% and a worst case of 1.78% of the 12-bit range, which the authors take as license to run the full benchmark on 59 dynamic trajectories from the extended BorealHDR dataset with stereo ORB-SLAM2. Their headline result is that the classical exposure method (AE50, a simple brightness-target controller) is the most dependable overall among eight tested methods, with a deep-reinforcement-learning method matching it on trajectory accuracy but failing more often. If the static error carries over to moving scenes, this framework turns auto-exposure evaluation into a reproducible, extensible offline experiment where any number of future methods can be compared without re-acquiring a single image.

What carries the argument

The load-bearing mechanism is the exposure emulator built on the remapping identity $$I_{\mathrm{target}} = f\!\left(\frac{\$\Delta$ t_{\mathrm{target}}}{\$\Delta$ t_{\mathrm{source}}} \cdot $f^{{-1}}$(I_{\mathrm{source}})\right),$$ which rescales an already-captured bracket to any desired exposure time using the camera response function $f$ and its inverse, estimated from multiple exposures of static scenes. Because the field brackets are captured while walking, the emulator deliberately does not interpolate between them; instead a selection rule named HIGHERNOSAT finds the two brackets that straddle the target exposure time, checks the saturation level of the higher bracket against a threshold $\alpha = 0.01$, and chooses the higher bracket when it is unsaturated (to keep signal-to-noise ratio high) and the lower bracket otherwise (to avoid clipping). This transforms the camera's exposure time from a hardware decision taken in the field into a controllable post-processing variable, which is what converts online, non-reproducible auto-exposure testing into an offline benchmark.

What would settle it

Record a bracketing burst while walking through a scene with strong parallax, and simultaneously capture ground-truth frames at intermediate exposure times with a second synchronized camera; if the emulator's RMSE on this moving sequence is substantially larger than the 0.21% to 1.78% range measured on static scenes, then every emulated trajectory in the eight-method comparison carries an unquantified photometric error and the ranking would need to be recomputed.

Watch

Extended reading notes

Core claim

The paper's central discovery is that exposure-time control, the active and environment-coupled part of a camera, can be moved into post-processing. Using the image formation model $I(x) = f(\Delta t \, V(x) \, E)$ and an estimated camera response function $f$, any bracketed image can be remapped to another exposure time by $$I_{\mathrm{target}} = f\!\left(\frac{\$\Delta$ t_{\mathrm{target}}}{\$\Delta$ t_{\mathrm{source}}} \cdot $f^{{-1}}$(I_{\mathrm{source}})\right).$$ The authors show, on five static test sequences of 1000 ground-truth images each spanning exposure times from $20\,\mu\mathrm{s}$ to $50\,\mathrm{ms}$, that this remapping reproduces real images with a median RMSE of 9 digital numbers (0.21%) and a worst case of 73 digital numbers (1.78%) of the 12-bit range, using a bracket-selection rule (HIGHERNOSAT) that picks the source exposure limiting saturation while preserving signal-to-noise advantage. With this emulator and the extended BorealHDR dataset, 59 trajectories covering 13.4 km with six repetitions of one trajectory across a single day, the authors benchmarked eight AE methods on ORB-SLAM2 and report that the classical AE50 method is the most dependable overall, with the most completed trajectories, the longest time before failure, and second-best trajectory accuracy, while a deep-reinforcement-learning method (DRL) achieves the lowest relative translation and rotation errors but loses reliability when deployed outside its training domain.

Load-bearing premise

The entire benchmark rests on the assumption that the same scene radiance reaches every image in a burst, even though the backpack was walking at about 2 km/h when the six brackets were captured; the paper validates the emulator only on static scenes and never quantifies how motion, parallax, or changing light between brackets degrades the 59 dynamic trajectories used to rank the eight auto-exposure methods.

Editorial extensions

If this is right

  • New auto-exposure methods can be added to the benchmark without any new data collection, because the emulator regenerates the exact image sequence the method would have commanded from stored brackets.
  • The comparison becomes a replayable public artifact, since the released code and the 59 trajectories let any group re-run the same evaluation and check the ranking.
  • For SLAM practitioners, the benchmark gives an evidence-backed design choice: the simple brightness-target method AE50 is the most dependable exposure controller for field deployment, while the learned DRL method is more accurate but less reliable outside its training domain.
  • Illumination becomes a controlled experimental variable: the Campus2 extension repeats the same 570 m loop six times in a single day, so methods face identical geometry under systematically changing light.
  • Each pairwise claim of superiority over the classical baseline is tied to a non-parametric significance test with a multiple-comparison correction, so the benchmark's conclusion is stated with an explicit confidence level.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial extension: because the emulator is validated only on static scenes, the margin between AE50 and DRL on trajectory accuracy is the most exposed ranking if motion and parallax between brackets inflate the error on dynamic scenes; a moving-scene validation would settle whether the ranking survives.
  • Editorial extension: the same emulation pipeline that benchmarks controllers could also train them, since 538,560 bracketed images can generate unlimited labelled exposure-response pairs offline, turning the dataset into a data engine for learning-based exposure control.
  • Editorial extension: adding the missing gain emulator could shift the benchmark's outcome, because the current setup forces methods to compensate for the absence of gain by requesting high exposure times, a behaviour the paper itself flags as a limitation.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 6 minor

Summary. This paper proposes an offline, replayable methodology for benchmarking camera auto-exposure (AE) methods, using a bracketed multi-exposure stereo dataset (BorealHDR) and a new extension where a single trajectory is repeated six times across a day. The core idea is an emulator that maps a "source" bracket to any target exposure time using the camera response function, with a heuristic (HIGHERNOSAT) for selecting the source bracket. The emulator is validated on static scenes (median RMSE 0.21%, maximum 1.78% of the 12-bit range). Using this emulator, the authors benchmark eight AE methods on 59 dynamic trajectories with a modified ORB-SLAM2, reporting feature statistics, relative trajectory error (RTE/RRE), robustness (time-to-failure and success counts), and a Mann-Whitney U-based statistical comparison. They conclude that the classical AE50 method remains the field's best performer overall, while the DRL method shows promise. The paper also documents the custom backpack platform, its hardware, electrical design, and lessons learned from field deployments.

Significance. If the central claim holds, the paper makes a valuable contribution: it would enable reproducible, offline comparison of any number of active exposure controllers on real field imagery, avoiding the cost and complexity of multi-camera or repeated-trajectory setups. The dataset extension (13.4 km, 59 trajectories, repeated trajectory under varying illumination) is a significant resource, and the static emulation validation is carefully designed, using 1000 ground-truth images, noise averaging, and a clear error metric. The detailed platform description and open-source code are also strengths. However, the paper's main benchmark conclusion rests on the assumption that the emulator is sufficiently accurate on the dynamic trajectories used for the AE comparison. That assumption is not adequately validated, and the SLAM evaluation itself is based on single runs of a nondeterministic algorithm. These issues currently weaken the support for the headline claims.

major comments (3)
  1. [Section III-A (Eq. 3), Section V-A, Section V-B] The emulation equation (Eq. 3) assumes that the same scene radiance reaches the sensor in the source and target brackets. The authors explicitly acknowledge in Section III-A that the brackets in BorealHDR are acquired while moving and are not spatially aligned, and they select the closest non-saturated bracket (Algorithm 1) rather than registering or interpolating. However, the only quantitative validation of the emulator (Section V-A) is performed on static scenes with ground-truth images. The dynamic 59-trajectory benchmark in Section V-B therefore inherits an unquantified error due to viewpoint motion between brackets, which includes parallax, object motion, and differing motion blur. This error is likely method-dependent: AE methods that request long exposure times will tend to draw from high-exposure source brackets, which exhibit more blur and larger viewpoint differences, whereas short-exposure methods will not. The limitation section (VI-A) mentions the absence of explicit motion blur modeling and gain, but it does not acknowledge the viewpoint-mismatch problem or its potential impact on the AE ranking. Without a dynamic validation—for example, by comparing emulated images to ground truth on a moving sequence, or at least a sensitivity analysis with respect to the choice of source bracket—the AE ranking cannot be considered supported.
  2. [Section V-B, Tables 1-3] The RTE, RRE, and failure-time results appear to be derived from a single run of ORB-SLAM2 per trajectory per AE method. In Section V-B3 the authors state that "ORB-SLAM2's non-deterministic nature can result in random failures," but no repeated runs are reported and no error bars or confidence intervals are provided. The point estimates in Table 1 and the statistical significance classifications in Table 3 are thus potentially dominated by seed-dependent random variation rather than by the AE method under test. The authors should either run the SLAM pipeline multiple times (e.g., 5-10 seeds) and report distributions, or use a deterministic variant, so that the comparisons are reproducible and statistically meaningful.
  3. [Section III-B4 and Table 3] The two-stage Mann-Whitney procedure used for Table 3 is not a standard approach for classifying methods as better/worse/equal. The first stage tests for any difference; only methods that pass this test proceed to a one-sided test for being better than AE50. Because the second test is conditional on the first, the Bonferroni correction applied in Eq. (9) does not control the overall false-discovery rate of the final three-way classification. In addition, the samples used in the test are per-trajectory means, and trajectories from the same location (e.g., the six Campus2 repeats) are likely correlated, which may inflate the effective sample size. The authors should use a single one-sided test per comparison (with a suitable multiplicity correction) or a proper closed-testing procedure, and account for the clustered structure of the data.
minor comments (6)
  1. [Abstract] The statement "we can emulate realistic images with a Root-Mean-Square Error (RMSE) below 1.78% compared to ground truth images" should be qualified as referring to the static validation only; the current phrasing overgeneralizes to dynamic scenes, which are not covered by the validation.
  2. [Section V-A] The text says the RMSE is "adjusted for camera noise," and Figure 6 labels the y-axis "RMSE above measured noise (DN)". Please make this explicit in the main text to avoid confusion about whether the 9 DN and 73 DN values include the sensor noise floor.
  3. [Section IV-C2] The phrase "as detailed in Section 2" should be replaced with a specific cross-reference, for example to Section IV-A2 where the RTK-GNSS receiver is described.
  4. [Section VI-A] The limitation paragraph states that the emulator "does not account for motion blur, with the only motion blur modeled being the inherent blur from the acquired image brackets." This is an understatement: because the source bracket is captured from a different viewpoint, the emulated image also contains the motion parallax and scene displacement of that source bracket. This should be acknowledged explicitly.
  5. [Section VII] There is a typo in the concluding sentence: "Our findings make us concludes that" should read "Our findings lead us to conclude that" or similar.
  6. [Section III-A] The notation for exposure times is inconsistent: sometimes Δt (e.g., Eq. (3)) and sometimes ΔT (e.g., ΔT_bracket). Please standardize to avoid confusion.

Circularity Check

1 steps flagged · score 6.0 of 10

The emulator's headline accuracy claim (RMSE below 1.78%) is an in-sample fit: the same five sequences are used both to compute the CRF and to evaluate the emulator, so the prediction is statistically forced.

  1. fitted input called prediction [Section IV-B (Emulator Calibration Sequences) and Section V-A (Emulation Static Validation)]
    "In total, five sequences comprising 1000 ground truth images were collected ... these calibration sequences were used to compute the Camera Response Function (CRF) ... [S]ection V-A: the performance of our image emulator is evaluated using five test sequences, each comprising 1000 ground truth images captured at exposure times ranging from 20 µs to 50 ms in both indoor and outdoor static scenes."

    The five static sequences described in Section IV-B are used to fit the CRF f(·) and its inverse, which is the key parameter in the emulation equation (Eq. 3: I_target = f((Δt_target/Δt_source) f^{-1}(I_source))). Section V-A then reports the emulator's RMSE as a 'validation' using the same five sequences, described with identical numbers and characteristics. The claimed sub-1.78% error is therefore a training/goodness-of-fit error, not an independent prediction on unseen scenes. The paper presents this in-sample fit as evidence that the emulator 'can emulate realistic images', and the subsequent AE benchmark inherits this unverified transfer claim.

full rationale

The single concrete circular step is the in-sample validation of the emulator: the CRF used in Eq. (3) is computed from the same five calibration sequences that are later called the five 'test' sequences for the RMSE claim. That makes the headline emulation-accuracy figure a fitted-input prediction rather than an independent measurement. The paper's central benchmark conclusion, however, does not reduce to this fit in the same way: the eight AE methods are run on the dynamic BorealHDR trajectories, which were not used to calibrate the CRF, and AE50 is an explicitly defined baseline rather than an optimized parameter. The self-citation to the authors' prior work [1] is load-bearing for the dataset and pipeline but is independently re-validated here and grounded in standard photometric equations cited to external works [36][37][38], so it does not constitute circularity by itself. The unquantified dynamic-scene error, acknowledged in Section VI-A ('the emulator does not account for motion blur'), is a validity limitation of the benchmark rather than a circular step; it should be weighed as correctness risk, not as additional circularity.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The framework adds no new physical entities. Its reliance on inherited photometric calibration and on the author-defined AE50 reference is explicit. The largest unquantified load is radiance invariance during motion, which affects every emulated frame in the benchmark.

free parameters (6)
  • Saturation threshold alpha = 0.01
    Used by HIGHERNOSAT to decide whether to use the higher bracketing exposure as source; a higher alpha would change which bracket is selected and hence the emulation error distribution. Chosen by the authors, not fitted from data.
  • Exposure bracket sets = BorealHDR: {1,2,4,8,16,32} ms; extension: {0.025,0.1,0.4,1.6,6.4,25.6} ms
    Design choices for dynamic range coverage and emulation frame rate; they affect achievable emulation accuracy and the motion-blur level present in the highest brackets.
  • AE baseline brightness targets = 30%, 50%, 70% of the 12-bit range
    The paper defines AE30, AE50, and AE70; AE50 is the 'classical' reference with which every other method is statistically compared, so choosing 50% as the reference is a hand-set decision that influences the headline comparison.
  • ORB-SLAM2 configuration = NFEATURES=3000, SCALEFACTOR=1.2, NLEVELS=13, INITHFAST=15, MINTHFAST=5; motion model disabled, all frames as…
    Fixed parameters for the single SLAM system used; they affect all downstream feature, RTE, RRE, and failure metrics. These are methodological choices not derived from data.
  • RTE and RRE evaluation window set = every integer w in [5, 50]
    The range of segment lengths over which relative errors are averaged in Eq. (6) and Eq. (8) was chosen by the authors; it weights how errors accumulate.
  • Feature coverage grid size = 20 by 20 cells
    Used to define 'filled' cells and coverage percentage in Section V-B1; changing grid resolution would change coverage values and rankings.
assumptions (4)
  • standard math Photometric model I(x) = f(Delta_t V(x) E) with an invertible CRF f
    Eqs. (2) and (3) assume a per-pixel radiometric model and an invertible, radiometrically calibrated camera response, inherited from [36], [37], and [38]. This is the mathematical foundation of the emulator.
  • domain assumption Scene radiance E is effectively constant between source and target brackets
    Eq. (3) remaps one bracket to another only if the same radiance reaches the sensor in both frames, but BorealHDR was captured while moving at about 2 km/h, so content and illumination shift between brackets. The paper mitigates by selecting rather than interpolating brackets but does not quantitatively validate this assumption on dynamic trajectories.
  • domain assumption Lidar-inertial reference trajectories are accurate enough to judge visual trajectories
    Section IV-C states the reference maps have median errors around 1% of trajectory length and are treated as adequate for comparison with visual SLAM outputs whose RTE values are roughly 18% to 32%.
  • domain assumption Feature metrics and a modified ORB-SLAM2 represent SLAM performance
    The benchmark evaluates exposure methods by feeding emulated images to a modified, non-real-time ORB-SLAM2; conclusions are implicitly limited to this pipeline and to feature-based localization.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reproducible Evaluation of Camera Auto-Exposure Methods in the Field: Platform, Benchmark and Lessons Learned." pith.science (2026). https://pith.science/paper/ZHP3IFMP

@misc{pith2026250618844,
  author       = {Pith},
  title        = {Pith review of: Reproducible Evaluation of Camera Auto-Exposure Methods in the Field: Platform, Benchmark and Lessons Learned},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZHP3IFMP}},
  note         = {Machine review of arXiv:2506.18844}
}
read the original abstract

Standard datasets often present limitations, particularly due to the fixed nature of input data sensors, which makes it difficult to compare methods that actively adjust sensor parameters to suit environmental conditions. This is the case with Automatic-Exposure (AE) methods, which rely on environmental factors to influence the image acquisition process. As a result, AE methods have traditionally been benchmarked in an online manner, rendering experiments non-reproducible. Building on our prior work, we propose a methodology that utilizes an emulator capable of generating images at any exposure time. This approach leverages BorealHDR, a unique multi-exposure stereo dataset, along with its new extension, in which data was acquired along a repeated trajectory at different times of the day to assess the impact of changing illumination. In total, BorealHDR covers 13.4 km over 59 trajectories in challenging lighting conditions. The dataset also includes lidar-inertial-odometry-based maps with pose estimation for each image frame, as well as Global Navigation Satellite System (GNSS) data for comparison. We demonstrate that by using images acquired at various exposure times, we can emulate realistic images with a Root-Mean-Square Error (RMSE) below 1.78% compared to ground truth images. Using this offline approach, we benchmarked eight AE methods, concluding that the classical AE method remains the field's best performer. To further support reproducibility, we provide in-depth details on the development of our backpack acquisition platform, including hardware, electrical components, and performance specifications. Additionally, we share valuable lessons learned from deploying the backpack over more than 25 km across various environments. Our code and dataset are available online at this link: https://github.com/norlab-ulaval/TFR24 BorealHDR

Figures

Figures reproduced from arXiv: 2506.18844 by the authors.

Figure 1
Figure 1. FIGURE 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 3
Figure 3. FIGURE 3 [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. FIGURE 4 [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figures from the paper (7 more)
Figure 5
Figure 5. Figure 5: FIGURE 5 [PITH_FULL_IMAGE:figures/full_fig_p010_5.png]
Figure 6
Figure 6. Figure 6: FIGURE 6 [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: FIGURE 7 [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]
Figure 8
Figure 8. Figure 8: FIGURE 8 [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: FIGURE 9 [PITH_FULL_IMAGE:figures/full_fig_p013_9.png]
Figure 10
Figure 10. Figure 10: FIGURE 10 [PITH_FULL_IMAGE:figures/full_fig_p015_10.png]
Figure 11
Figure 11. Figure 11: FIGURE 11 [PITH_FULL_IMAGE:figures/full_fig_p016_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 45 canonical work pages

  1. [1]

    Exposing the Unseen: Exposure Time Emulation for Offline Benchmarking of Vision Algorithms

    O. Gamache, J.-M. Fortin, M. Boxan, M. Vaidis, F. Pomerleau, and P. Gigu `ere, “Exposing the Un- seen: Exposure Time Emulation for Offline Bench- marking of Vision Algorithms,”arXiv preprint arXiv:2309.13139, 2023

    work page Pith review arXiv 2023

  2. [2]

    Towards Automated System and Exper- iment Reproduction in Robotics,

    F. Lier et al., “Towards Automated System and Exper- iment Reproduction in Robotics,” in2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), IEEE, 2016, pp. 3298–3305

    work page 2016

  3. [3]

    Vision meets robotics: The KITTI dataset,

    A. Geiger, P. Lenz, C. Stiller, and R. Urtasun, “Vision meets robotics: The KITTI dataset,”The Interna- tional Journal of Robotics Research, vol. 32, no. 11, pp. 1231–1237, 2013

    work page 2013

  4. [4]

    1 Year, 1000km: The Oxford RobotCar Dataset,

    W. Maddern, G. Pascoe, C. Linegar, and P. Newman, “1 Year, 1000km: The Oxford RobotCar Dataset,”The International Journal of Robotics Research, vol. 36, no. 1, pp. 3–15, 2017

    work page 2017

  5. [5]

    Simultaneous Localization and Mapping Related Datasets: A Comprehensive Survey

    Y . Liu, Y . Fu, F. Chen, B. Goossens, W. Tao, and H. Zhao, “Simultaneous Localization and Mapping Related Datasets: A Comprehensive Survey,”arXiv preprint arXiv:2102.04036, 2021

    work page Pith review arXiv 2021

  6. [6]

    Survey of Monocular SLAM Algorithms in Natural Environments,

    G. Chahine and C. Pradalier, “Survey of Monocular SLAM Algorithms in Natural Environments,” inCon- ference on Computer and Robot Vision (CRV), 2018

    work page 2018

  7. [7]

    Learned Camera Gain and Exposure Control for Improved Visual Feature Detection and Matching,

    J. Tomasi, B. Wagstaff, S. L. Waslander, and J. Kelly, “Learned Camera Gain and Exposure Control for Improved Visual Feature Detection and Matching,” IEEE Robotics and Automation Letters (RA-L), vol. 6, no. 2, pp. 2028–2035, 2021

    work page 2021

  8. [8]

    Kilometer-scale autonomous navi- gation in subarctic forests: Challenges and lessons learned,

    D. Baril et al., “Kilometer-scale autonomous navi- gation in subarctic forests: Challenges and lessons learned,”Field Robotics, vol. 2, no. 1, pp. 1628–1660, Mar. 2022

    work page 2022

Show all 46 references
  1. [9]

    Active exposure control for robust visual odometry in HDR environments,

    Z. Zhang, C. Forster, and D. Scaramuzza, “Active exposure control for robust visual odometry in HDR environments,” inIEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 3894– 3901

    2017

  2. [10]

    Proactive Cam- era Attribute Control Using Bayesian Optimization for Illumination-Resilient Visual Navigation,

    J. Kim, Y . Cho, and A. Kim, “Proactive Cam- era Attribute Control Using Bayesian Optimization for Illumination-Resilient Visual Navigation,”IEEE Transactions on Robotics (T-RO), vol. 36, pp. 1256– 1271, 2020

    2020

  3. [11]

    Auto-Exposure Al- gorithm for Enhanced Mobile Robot Localization in Challenging Light Conditions,

    M.-A. B ´egin and I. Hunter, “Auto-Exposure Al- gorithm for Enhanced Mobile Robot Localization in Challenging Light Conditions,”Sensors, vol. 22, p. 835, 3 2022

    2022

  4. [12]

    Camera Exposure Control for Robust Robot Vision with Noise-Aware Image Quality Assessment,

    U. Shin, J. Park, G. Shim, F. Rameau, and I. S. Kweon, “Camera Exposure Control for Robust Robot Vision with Noise-Aware Image Quality Assessment,” inIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019, pp. 1165–1172

    2019

  5. [13]

    Gradient-based Camera Exposure Control for Outdoor Mobile Platforms,

    I. Shim, T. H. Oh, J. Y . Lee, J. Choi, D. G. Choi, and I. S. Kweon, “Gradient-based Camera Exposure Control for Outdoor Mobile Platforms,”IEEE Trans- actions on Circuits and Systems for Video Technology, vol. 29, pp. 1569–1583, 6 Jun. 2019

    2019

  6. [14]

    Fibonacci Expo- sure Bracketing for High Dynamic Range Imaging,

    M. Gupta, D. Iso, and S. K. Nayar, “Fibonacci Expo- sure Bracketing for High Dynamic Range Imaging,” inIEEE International Conference on Computer Vision (ICCV), 2013

    2013

  7. [15]

    Benefit of large field-of-view cameras for vi- sual odometry,

    Z. Zhang, H. Rebecq, C. Forster, and D. Scara- muzza, “Benefit of large field-of-view cameras for vi- sual odometry,” inIEEE International Conference on Robotics and Automation (ICRA), 2016, pp. 801–808

    2016

  8. [16]

    Learning-based image enhancement for visual odometry in challenging HDR environ- ments,

    R. Gomez-Ojeda, Z. Zhang, J. Gonzalez-Jimenez, and D. Scaramuzza, “Learning-based image enhancement for visual odometry in challenging HDR environ- ments,” inIEEE International Conference on Robotics and Automation (ICRA), 2018, pp. 805–811

    2018

  9. [17]

    Sim2Real in robotics and automation: Applications and challenges,

    S. H ¨ofer et al., “Sim2Real in robotics and automation: Applications and challenges,”IEEE transactions on automation science and engineering, vol. 18, no. 2, pp. 398–400, 2021

    2021

  10. [18]

    Gradient-Based Auto-Exposure Control Applied to a Self-Driving Car,

    I. Mehta, M. Tang, and T. D. Barfoot, “Gradient-Based Auto-Exposure Control Applied to a Self-Driving Car,” inConference on Computer and Robot Vision (CRV), 2020, pp. 166–173

    2020

  11. [19]

    Auto- mated camera-exposure control for robust localization in varying illumination environments,

    Y . Wang, H. Chen, S. Zhang, and W. Lu, “Auto- mated camera-exposure control for robust localization in varying illumination environments,”Autonomous Robots, vol. 46, no. 4, pp. 515–534, 2022

    2022

  12. [20]

    Camera Attributes Control for Visual Odome- try With Motion Blur Awareness,

    B. Han, Y . Lin, Y . Dong, H. Wang, T. Zhang, and C. Liang, “Camera Attributes Control for Visual Odome- try With Motion Blur Awareness,”IEEE/ASME Trans- actions on Mechatronics (TMECH), vol. 28, no. 4, pp. 2225–2235, 2023

    2023

  13. [21]

    Effi- cient Camera Exposure Control for Visual Odometry via Deep Reinforcement Learning,

    S. Zhang, J. He, Y . Zhu, J. Wu, and J. Yuan, “Effi- cient Camera Exposure Control for Visual Odometry via Deep Reinforcement Learning,”arXiv preprint arXiv:2408.17005, 2024

    2024 arXiv

  14. [22]

    Kitti-360: A novel dataset and benchmarks for urban scene understanding in 2d and 3d,

    Y . Liao, J. Xie, and A. Geiger, “Kitti-360: A novel dataset and benchmarks for urban scene understanding in 2d and 3d,”IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 45, no. 3, pp. 3292– 3310, 2022

    2022

  15. [23]

    The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset,

    D. Barnes, M. Gadd, P. Murcutt, P. Newman, and I. Posner, “The Oxford Radar RobotCar Dataset: A Radar Extension to the Oxford RobotCar Dataset,” VOLUME , 17 Author et al.: inIEEE International Conference on Robotics and Automation (ICRA), May 2020, pp. 6433–6438

    2020

  16. [24]

    Learning visual feature descriptors for dynamic lighting con- ditions,

    N. Carlevaris-Bianco and R. M. Eustice, “Learning visual feature descriptors for dynamic lighting con- ditions,” inIEEE/RSJ International Conference on Intelligent Robots and Systems, 2014, pp. 2769–2776

    2014

  17. [25]

    The UMA-VI dataset: Vi- sual–inertial odometry in low-textured and dynamic illumination environments,

    D. Zu ˜niga-No¨el, A. Jaenal, R. Gomez-Ojeda, and J. Gonzalez-Jimenez, “The UMA-VI dataset: Vi- sual–inertial odometry in low-textured and dynamic illumination environments,”The International Journal of Robotics Research, vol. 39, pp. 1052–1060, 9 Aug. 2020

    2020

  18. [26]

    TartanDrive: A Large-Scale Dataset for Learning Off-Road Dynam- ics Models,

    S. Triest, M. Sivaprakasam, S. J. Wang, W. Wang, A. M. Johnson, and S. Scherer, “TartanDrive: A Large-Scale Dataset for Learning Off-Road Dynam- ics Models,” inIEEE International Conference on Robotics and Automation (ICRA), 2022, pp. 2546– 2552

    2022

  19. [27]

    FinnForest dataset: A forest landscape for visual SLAM,

    I. Ali et al., “FinnForest dataset: A forest landscape for visual SLAM,”Robotics and Autonomous Systems, vol. 132, p. 103 610, Oct. 2020

    2020

  20. [28]

    The visual–inertial canoe dataset,

    M. Miller, S.-J. Chung, and S. Hutchinson, “The visual–inertial canoe dataset,”The International Jour- nal of Robotics Research, vol. 37, no. 1, pp. 13–20, 2018

    2018

  21. [29]

    Wild-places: A large-scale dataset for lidar place recognition in unstructured natural environments,

    J. Knights, K. Vidanapathirana, M. Ramezani, S. Srid- haran, C. Fookes, and P. Moghadam, “Wild-places: A large-scale dataset for lidar place recognition in unstructured natural environments,” in2023 IEEE international conference on robotics and automation (ICRA), IEEE, 2023, p...

    2023

  22. [30]

    A low- drift and real-time localisation and mapping method for handheld hemispherical view lidar-imu integra- tion system,

    X. Duan, Q. Hu, P. Zhao, F. Yu, and M. Ai, “A low- drift and real-time localisation and mapping method for handheld hemispherical view lidar-imu integra- tion system,”The Photogrammetric Record, vol. 38, no. 182, pp. 176–196, 2023

    2023

  23. [31]

    PolyU-BPCoMa: A dataset and bench- mark towards mobile colorized mapping using a back- pack multisensorial system,

    W. Shi et al., “PolyU-BPCoMa: A dataset and bench- mark towards mobile colorized mapping using a back- pack multisensorial system,”International Journal of Applied Earth Observation and Geoinformation, vol. 112, p. 102 962, 2022

    2022

  24. [32]

    Low-drift odometry, mapping and ground segmentation using a backpack lidar system,

    P. Chen, W. Shi, S. Bao, M. Wang, W. Fan, and H. Xiang, “Low-drift odometry, mapping and ground segmentation using a backpack lidar system,”IEEE Robotics and Automation Letters, vol. 6, no. 4, pp. 7285–7292, 2021

    2021

  25. [33]

    A systematic mapping framework for backpack mobile mapping system in common monotonous environ- ments,

    S. Bao, W. Shi, P. Chen, H. Xiang, and Y . Yu, “A systematic mapping framework for backpack mobile mapping system in common monotonous environ- ments,”Measurement, vol. 197, p. 111 243, 2022

    2022

  26. [34]

    Backpack LiDAR-based SLAM with multiple ground constraints for multistory indoor mapping,

    B. Zhou, H. Mo, S. Tang, X. Zhang, and Q. Li, “Backpack LiDAR-based SLAM with multiple ground constraints for multistory indoor mapping,”IEEE Transactions on Geoscience and Remote Sensing, vol. 61, pp. 1–16, 2023

    2023

  27. [35]

    A mathematical theory of commu- nication,

    C. E. Shannon, “A mathematical theory of commu- nication,”The Bell system technical journal, vol. 27, no. 3, pp. 379–423, 1948

    1948

  28. [36]

    Online photometric calibration of auto exposure video for realtime visual odometry and slam,

    P. Bergmann, R. Wang, and D. Cremers, “Online photometric calibration of auto exposure video for realtime visual odometry and slam,”IEEE Robotics and Automation Letters, vol. 3, pp. 627–634, 2 Apr. 2018

    2018

  29. [37]

    Determining the camera response from images: What is knowable?

    M. Grossberg and S. Nayar, “Determining the camera response from images: What is knowable?”IEEE Transactions on Pattern Analysis and Machine Intel- ligence, vol. 25, pp. 1455–1467, 11 Nov. 2003

    2003

  30. [38]

    A photo- metrically calibrated benchmark for monocular visual odometry,

    J. Engel, V . Usenko, and D. Cremers, “A photo- metrically calibrated benchmark for monocular visual odometry,” inarXiv:1607.02555, Jul. 2016

    2016 arXiv

  31. [39]

    A benchmark for the evaluation of rgb-d slam systems,

    J. Sturm, N. Engelhard, F. Endres, W. Burgard, and D. Cremers, “A benchmark for the evaluation of rgb-d slam systems,” inIEEE/RSJ international conference on intelligent robots and systems, 2012, pp. 573–580

    2012

  32. [40]

    Deep introspective SLAM: Deep reinforcement learning based approach to avoid tracking failure in visual SLAM,

    K. Naveed, M. L. Anjum, W. Hussain, and D. Lee, “Deep introspective SLAM: Deep reinforcement learning based approach to avoid tracking failure in visual SLAM,”Autonomous Robots, vol. 46, no. 6, pp. 705–724, 2022

    2022

  33. [41]

    On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other,

    H. B. Mann and D. R. Whitney, “On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other,”The annals of mathematical statistics, pp. 50–60, 1947

    1947

  34. [42]

    Teoria statistica delle classi e calcolo delle probabilit `a,

    C. Bonferroni, “Teoria statistica delle classi e calcolo delle probabilit `a,”Pubblicazioni del R istituto superi- ore di scienze economiche e commericiali di firenze, vol. 8, pp. 3–62, 1936

    1936

  35. [43]

    Tar- getless Calibration of LiDAR-IMU System Based on Continuous-time Batch Estimation,

    J. Lv, J. Xu, K. Hu, Y . Liu, and X. Zuo, “Tar- getless Calibration of LiDAR-IMU System Based on Continuous-time Batch Estimation,” in2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020, pp. 9968–9975

    2020

  36. [44]

    Gravity- constrained point cloud registration,

    V . Kubelka, M. Vaidis, and F. Pomerleau, “Gravity- constrained point cloud registration,” inIEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2022, pp. 4873–4879

    2022

  37. [45]

    Orb-slam2: An open- source slam system for monocular, stereo, and rgb- d cameras,

    R. Mur-Artal and J. D. Tard ´os, “Orb-slam2: An open- source slam system for monocular, stereo, and rgb- d cameras,”IEEE transactions on robotics, vol. 33, no. 5, pp. 1255–1262, 2017

    2017

  38. [46]

    Adaptive Risk Minimiza- tion: Learning to Adapt to Domain Shift,

    M. Zhang, H. Marklund, N. Dhawan, A. Gupta, S. Levine, and C. Finn, “Adaptive Risk Minimiza- tion: Learning to Adapt to Domain Shift,”Advances in Neural Information Processing Systems, vol. 34, pp. 23 664–23 678, 2021. 18 VOLUME , <Society logo(s) and publication title will ap...

    2021

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.