Pith. sign in

REVIEW 4 major objections 5 minor 24 references

Automated vision-based assistance tools in bronchoscopy: stenosis severity estimation

T0 review · 4 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read The paper proposes a fully automated bronchoscopy-only pipeline that measures subglottic stenosis severity from a single keyframe, using illumination decline to track the lumen and reconstruct the airway in 3D.

desk verdict A genuinely useful first benchmark and a clever illumination-decline pipeline for subglottic stenosis, but the evaluation has a train/eval overlap and an unvalidated lumen segmentation that need scrutiny. read the letter →

arxiv 2505.05136 v1 pith:7KKDFCGO submitted 2025-05-08 cs.CV

Clara Tomasini , Javier Rodriguez-Puigvert , Dinora Polanco , Manuel Viñuales , Luis Riazuelo , Ana Cristina Murillo This is my paper · ORCID
classification cs.CV
keywords bronchoscopysubglotticstenosisseverityestimationilluminationdeclinesingle-viewdepthairwaylumensegmentationIoUtracking3Dreconstruction
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

This paper aims to turn a routine bronchoscopy video into an objective measurement of subglottic stenosis severity, replacing the subjective visual inspection that prior work has repeatedly shown to be unreliable. The proposed two-step pipeline follows the darkest region of the image—the airway lumen—until the scope passes the vocal cords, then builds a 3D model of the airway from a single keyframe and computes the percentage of obstruction. The authors claim this is the first automated method for this measurement that needs no traversal of the stenosed region, no CT scan, and no additional hardware. They support the claim with a new public dataset of real procedures, reporting stenosis estimates consistent with CT ground truth and expert judgment, and repeatable across repeated procedures on the same patient.

What carries the argument

The load-bearing mechanism is the photometric inverse-square law: because the bronchoscope's light source and camera are co-located, image intensity $I$ falls with distance $d$ roughly as $I \approx (1/d^2)^{1/\gamma}$, so the darkest area marks the deepest visible point, the lumen. That cue drives both pipeline stages: an IoU tracker follows the darkest segment until its shape changes abruptly at the vocal cords, and a single-view self-supervised depth model trained on the same illumination-decline signal produces the 3D airway points $X_{3D}$ from the selected keyframe. Stenosed points are isolated as the darkest 3D set $X_{3D}^{\text{stenosis}} = \{X \in X_{3D} \mid X \in P\}$, where $P$ is the plane fitted to them, and the healthy reference area is the maximum over planes $Q_k$ perpendicular to the camera: $k = \arg\max_k A_k$. The ratio of these two areas or of fitted circle diameters yields the stenosis index. Keyframe selection and 3D geometry are therefore two uses of the same physical signal.

What would settle it

Run the pipeline on the healthy patients in the dataset (G, H, I), who have no stenosis: a correct method should produce near-zero stenosis indices, so a large reported $PSA$ or $PSD$ on those sequences would show that the darkest-region assumption is measuring shadow or tissue rather than lumen. A second check would be a phantom airway with known obstruction diameters placed at known distances from the camera; systematic deviation of estimated $PSA/PSD$ from the known geometry would falsify the claim that the single-frame 3D model preserves airway proportions.

Watch

Extended reading notes

Core claim

The paper's central discovery, stated on its own terms, is that illumination decline in endoscopy carries enough geometric information to automate subglottic stenosis severity estimation from a single bronchoscopy frame. By segmenting and tracking the darkest image patch with an IoU tracker, the pipeline detects when the camera has just passed the vocal cords; the frame where tracking is lost becomes the keyframe. A single-view self-supervised depth model reconstructs the airway in 3D, the stenosed region is delineated as the darkest set of 3D points and intersected with a plane, and the healthy reference is taken as the largest plane section between the vocal cords and the stenosis. The stenosis index is reported both as an area ratio, $PSA = (1 - A_{\text{stenosis}}/A_k)\times 100$, and a diameter ratio, $PSD = (1 - d_{\text{stenosis}}/d_k)\times 100$. On the released dataset the pipeline selects correct keyframes for all 16 sequences, achieves mean absolute errors of 4.84% for $PSA$ and 6.70% for $PSD$, and shows same-patient differences of 1.79% for $PSA$ and 5.07% for $PSD$, compared with 7.5% average variation between expert estimates; the authors also note a slight underestimation of the stenosis index and a healthy-reference location that differs from the clinical Myer-Cotton grading.

Load-bearing premise

The method assumes that just after the tracked darkest patch changes shape the camera has passed through the vocal cords, and that in that keyframe the darkest image region is the stenosed airway opening rather than shadow or tissue, with the 3D shape computed from that single frame accurate enough that the area and diameter ratios match the real airway.

Editorial extensions

If this is right

  • Stenosis severity can be measured during the same bronchoscopy pass used for exploration, without threading the scope through the narrowing, which shortens the procedure and removes the need for a CT scan for monitoring.
  • An automated, reproducible number replaces the subjective visual grade; the paper cites prior work in which physicians misclassified severity in 53% of cases, mostly by underestimation.
  • Because only a single keyframe is needed, the measurement can run in parallel with the clinician's assessment, taking about 7 seconds per case on a current GPU.
  • The healthy reference is measured between the vocal cords and the stenosis rather than in the trachea beyond it, so a direct mapping to the standard Myer-Cotton grade still requires calibration with expert-supervised data.
  • The released benchmark gives other research groups the first public data on which to train and compare automated stenosis estimators.

Reading between the lines

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

  • The paper leaves implicit that the same illumination-decline cue could extend to other endoluminal strictures, such as tracheal or esophageal narrowing, wherever a co-located light and camera produce the same intensity-distance relation.
  • The paper does not draw the consequence that combining the scale-invariant ratio with the known bronchoscope diameter would yield absolute lumen diameters and a mapping onto the Myer-Cotton grading scale.
  • A testable extension the paper does not pursue is to use the albedo and normal maps the depth model already outputs to distinguish a true lumen from shadowed tissue, which would harden the darkest-equals-lumen premise.
  • The same-patient repeatability suggests a longitudinal monitoring tool, but the paper only shows two sequences per patient, so sensitivity to genuine stenosis progression over time is not established.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The manuscript proposes a fully automated pipeline for estimating subglottic stenosis severity from bronchoscopy video. The method segments the darkest image region (intensity < 50), tracks it with an IoU tracker to detect when the bronchoscope passes the vocal cords, selects that frame as the keyframe, and applies LightDepth single-view depth reconstruction to measure the stenotic area and diameter ratios against a maximum reference area between the vocal cords and the stenosis. The authors contribute a new dataset of 16 real bronchoscopy videos from 11 patients with CT-based and expert-based references, and report a PSA MAE of 4.84% and a PSD MAE of 6.70%, with perfect keyframe selection on their method. They claim this is the first automated, robust SGS measurement from bronchoscopy images.

Significance. If the results hold, the work would provide a practical assistive tool that reduces subjectivity and CT radiation exposure, and the released benchmark would be a useful community resource. The use of external CT and expert references, repeatability analysis across repeated procedures, and comparison against keyframe-selection baselines are strengths. However, the current evidence is limited by the small cohort, the training/evaluation overlap for sequence 1, and the absence of validation of the intensity-mask segmentation that underlies the measurement; these gaps currently weaken the central claim of robustness. The paper also ships a public dataset, which is a tangible contribution to a data-scarce area.

major comments (4)
  1. [4.1] LightDepth is trained on 2800 frames from sequence 1 and that same sequence is retained in the stenosis evaluation (Section 4.1, 'This sequence is used for LightDepth training, but is kept to evaluate...'). Sequence 1 also has CT ground truth. Because the depth model is evaluated on training data, the reported MAE for that sequence and the aggregate MAE may be optimistically biased, and the claim of consistency with CT is weakened. Please report results separately for sequence 1 versus held-out sequences, or retrain LightDepth on a split that excludes sequence 1 and re-run the evaluation.
  2. [3.3, Eqs. (1)-(3)] The stenosed area Astenosis is defined by the binary intensity threshold (grayscale < 50) of Section 3.2, and this mask propagates directly into PSA and PSD via Eqs. (1)-(3). The paper validates keyframe selection (Correct Keyframes) but never validates the lumen mask against manual delineations or any independent segmentation. If the darkest region is a shadow or a mucosal fold, the reported agreement with CT and expert references could be coincidental. Please provide a segmentation-level validation (e.g., Dice against expert lumen annotations on a subset of keyframes) and a sensitivity analysis of PSA/PSD to the intensity threshold.
  3. [4.2, Table 2] The MAE values are reported without confidence intervals or per-sequence error distributions, and the paired t-tests (p=0.6039, p=0.2766) are underpowered with N=5 CT and N=12 expert comparisons; 'no statistically significant difference' is not evidence of agreement. Please report Bland-Altman limits of agreement and bootstrap confidence intervals for the MAE, and discuss the effect size relative to clinically meaningful differences.
  4. [5, Fig. 4] The healthy reference area Ak is measured between the vocal cords and the stenosis, whereas the CT ground truth is defined at the trachea's widest point below the stenosis and expert estimates follow the Myer-Cotton convention. This is acknowledged as a limitation, but it means the reported agreement in Table 2 is between two different anatomical quantities; the 'slight underestimation' could be entirely explained by this location mismatch. Please quantify the expected bias by computing both reference locations on the CT volumes for the five CT cases, or explicitly report the Myer-Cotton-equivalent index.
minor comments (5)
  1. [1] The name 'LigthDepth' in the Introduction (Section 1) is a typo; it should be 'LightDepth' as used in the rest of the manuscript.
  2. [Table 2] The table mixes comma and period as decimal separators (e.g., '37,5' vs '4.84'); please use a consistent decimal separator throughout.
  3. [3.3] The notation in Eq. (1) and the surrounding text, particularly 'X∋D', is confusing; please clarify what D denotes and how the set X3Dstenosis is obtained from the plane intersection.
  4. [3.2] The sentence 'When the tracker stops...' contains a stray closing parenthesis and could be rephrased for clarity: remove the extra ')' and state directly that tracking loss indicates the camera has passed the vocal cords.
  5. [Abstract] The claim 'No public methods or datasets are currently available' is stronger than warranted given the semi-automatic methods cited in [11,12]; consider specifying 'no publicly available automated method or benchmark'.

Circularity Check

1 steps flagged · score 5.0 of 10

LightDepth is trained on sequence 1 and sequence 1 is then included in the stenosis evaluation, making that sequence's PSA/PSD an in-sample fit; the rest of the pipeline is anchored by external CT and expert references.

  1. fitted input called prediction [Section 4.1, '3D Reconstruction model training']
    "We train LightDepth on 2800 frames from sequence 1 of the Subglottic Stenosis dataset, covering various anatomical regions, including the subglottic and tracheal areas, with the corresponding geometric and photometric calibration of the bronchoscope. This sequence is used for LightDepth training, but is kept to evaluate the quality of stenosis estimation, as it includes scarce CT ground-truth data and that errors in this sequence are similar than in the rest."

    LightDepth is the module that produces the depth map from which Astenosis and Ak in Eqs. (1)-(3) are computed. Training it on 2800 frames from sequence 1 means the depth estimates for the sequence-1 keyframe are fitted to that sequence's photometric data, not predicted independently. Sequence 1 is then included in all 16 evaluated sequences, and its CT ground truth contributes to the MAE reported in Table 2. The paper even justifies keeping it because its errors are similar to the rest, i.e., the in-sample result is used as evidence of accuracy. For that sequence the 'prediction' reduces to a fit; the held-out sequences carry the independent evidence.

full rationale

The paper's internal derivation is mostly self-contained: keyframe selection uses a photometric prior and IoU tracking; stenosis measurement back-projects the segmented dark region through a single-view depth map; and the results are checked against external CT and expert estimates on a newly released dataset. The clearest circular element is in Section 4.1: LightDepth, the module that produces the depth geometry used in Eqs. (1)-(3), is trained on 2800 frames from sequence 1, and sequence 1 is then kept in the evaluation and contributes its CT ground truth to the reported MAE. For the sequence-1 keyframe the depth is an in-sample fit, so that particular PSA/PSD is not a genuine prediction. The paper's own sentence confirms the overlap. Held-out sequences 2-16 and the external CT/expert comparisons provide independent evidence, so the central claim does not collapse entirely. No uniqueness theorem or ansatz is smuggled via self-citation: LightDepth is a published method with its own external evaluation, and the paper additionally compares against DepthAnything and COLMAP. A remaining non-circular risk is that the intensity-threshold mask is never validated against manual lumen annotations; if the darkest region is a shadow or tissue, Eqs. (1)-(3) would measure the artifact, and the external CT/expert anchoring would only detect this in aggregate. Overall the paper has one in-sample evaluation step rather than a fully circular derivation.

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

The approach contributes a new application and dataset, but the central claim rests on several domain assumptions and training choices. The main costs are the fixed intensity threshold, tracking parameters, LightDepth training hyperparameters that include a test sequence, and the assumption that darkness corresponds to lumen and depth. No new physical entity is invented.

free parameters (6)
  • Darkness segmentation threshold = 50 (grayscale intensity)
    Binary threshold chosen by visual assessment on colonoscopy data following illumination decline; it controls which pixels are considered lumen or stenosis and therefore affects keyframe detection and stenosed area.
  • Minimum IoU for tracking = 50%
    IoU threshold for considering a tracked segment still matched; a different threshold changes when the tracker is lost and hence which keyframe is selected.
  • Tracking loss frame count = 25 frames
    Number of consecutive failed IoU matches before the tracker is declared lost; directly sets when the keyframe is reached.
  • LightDepth training epochs = 5
    Training schedule for the depth model; no evidence is shown that this schedule is stable, and training includes sequence 1 used in evaluation.
  • LightDepth learning rate = 1e-4
    Chosen optimizer setting for LightDepth training; affects the geometry learned by the depth model.
  • Depth smoothness and specularity loss weights = lambda_s = 0.1, lambda_sp = 1
    Regularization weights for LightDepth training; they influence the albedo and depth estimates used for stenosis measurement.
assumptions (6)
  • domain assumption Co-location of camera and light source in bronchoscopy makes image intensity decay with distance according to the inverse-square law and gamma correction.
    Used in Sections 3.1 and 3.3 to justify both darkest-region tracking and LightDepth's photometric reconstruction.
  • domain assumption The dark region in the keyframe corresponds to the stenosed lumen and not to shadow, secretions, or other anatomy.
    Used in Section 3.3 to delineate the stenosed region; not verified against independent annotations.
  • domain assumption LightDepth's single-view up-to-scale depth estimates are accurate enough in the subglottis for area and diameter ratios to be meaningful.
    The paper relies on the ICCV LightDepth model rather than validating depth accuracy on this dataset; only indirect stenosis error is reported.
  • domain assumption Loss of the IoU-tracked darkest segment marks passage of the vocal cords and entrance to the subglottis.
    Section 3.2 defines this event as the keyframe; no independent anatomical localization is provided.
  • domain assumption CT-derived ground-truth areas and expert diameter estimates are valid references for stenosis severity.
    Used as ground truth in Table 2; expert estimates themselves are acknowledged as subjective in the introduction.
  • ad hoc to paper The healthy reference area should be measured between the vocal cords and the stenosis rather than in the trachea below the stenosis.
    The paper acknowledges this differs from the clinically used Myer-Cotton classification in the Conclusions, which may explain the reported systematic underestimation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Automated vision-based assistance tools in bronchoscopy: stenosis severity estimation." pith.science (2026). https://pith.science/paper/7KKDFCGO

@misc{pith2026250505136,
  author       = {Pith},
  title        = {Pith review of: Automated vision-based assistance tools in bronchoscopy: stenosis severity estimation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7KKDFCGO}},
  note         = {Machine review of arXiv:2505.05136}
}
read the original abstract

Purpose: Subglottic stenosis refers to the narrowing of the subglottis, the airway between the vocal cords and the trachea. Its severity is typically evaluated by estimating the percentage of obstructed airway. This estimation can be obtained from CT data or through visual inspection by experts exploring the region. However, visual inspections are inherently subjective, leading to less consistent and robust diagnoses. No public methods or datasets are currently available for automated evaluation of this condition from bronchoscopy video. Methods: We propose a pipeline for automated subglottic stenosis severity estimation during the bronchoscopy exploration, without requiring the physician to traverse the stenosed region. Our approach exploits the physical effect of illumination decline in endoscopy to segment and track the lumen and obtain a 3D model of the airway. This 3D model is obtained from a single frame and is used to measure the airway narrowing. Results: Our pipeline is the first to enable automated and robust subglottic stenosis severity measurement using bronchoscopy images. The results show consistency with ground-truth estimations from CT scans and expert estimations, and reliable repeatability across multiple estimations on the same patient. Our evaluation is performed on our new Subglottic Stenosis Dataset of real bronchoscopy procedures data. Conclusion: We demonstrate how to automate evaluation of subglottic stenosis severity using only bronchoscopy. Our approach can assist with and shorten diagnosis and monitoring procedures, with automated and repeatable estimations and less exploration time, and save radiation exposure to patients as no CT is required. Additionally, we release the first public benchmark for subglottic stenosis severity assessment.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

24 extracted references · 22 canonical work pages

  1. [1]

    what do bronchoscopists do? Annals of the American Thoracic Society 12(1), 85–90 (2015) 10

    Begnaud, A., Connett, J.E., Harwood, E.M., Jantz, M.A., Mehta, H.J.: Measuring central airway obstruction. what do bronchoscopists do? Annals of the American Thoracic Society 12(1), 85–90 (2015) 10

    work page 2015

  2. [2]

    Interactive cardiovascular and thoracic surgery 16(5), 655–660 (2013)

    Murgu, S., Colt, H.: Subjective assessment using still bronchoscopic images mis- classifies airway narrowing in laryngotracheal stenosis. Interactive cardiovascular and thoracic surgery 16(5), 655–660 (2013)

    work page 2013

  3. [3]

    In: IEEE/CVF Int

    Rodr´ ıguez-Puigvert, J., Batlle, V.M., Montiel, J.M.M., Martinez-Cantin, R., Fua, P., Tard´ os, J.D., Civera, J.: Lightdepth: Single-view depth self-supervision from illumination decline. In: IEEE/CVF Int. Conf. on Computer Vision (ICCV), pp. 21273–21283 (2023)

    work page 2023

  4. [4]

    Annals of Otology, Rhinology & Laryngology 125(10), 815–822 (2016)

    Sharma, G.K., Chin Loy, A., Su, E., Jing, J., Chen, Z., Wong, B.J., Verma, S.: Quantitative evaluation of adult subglottic stenosis using intraoperative long- range optical coherence tomography. Annals of Otology, Rhinology & Laryngology 125(10), 815–822 (2016)

    work page 2016

  5. [5]

    International Journal of Computer Assisted Radiology and Surgery 18(4), 707–713 (2023)

    Banach, A., Naito, M., King, F., Masaki, F., Tsukada, H., Hata, N.: Computer- based airway stenosis quantification from bronchoscopic images: preliminary results from a feasibility trial. International Journal of Computer Assisted Radiology and Surgery 18(4), 707–713 (2023)

    work page 2023

  6. [6]

    International journal of computer assisted radiology and surgery 10, 935–945 (2015)

    S´ anchez, C., Bernal, J., S´ anchez, F.J., Diez, M., Rosell, A., Gil, D.: Toward online quantification of tracheal stenosis from videobronchoscopy. International journal of computer assisted radiology and surgery 10, 935–945 (2015)

    work page 2015

  7. [7]

    In: Medical Imaging 2023: Image Processing, vol

    Keuth, R., Heinrich, M., Eichenlaub, M., Himstedt, M.: Weakly supervised air- way orifice segmentation in video bronchoscopy. In: Medical Imaging 2023: Image Processing, vol. 12464, pp. 58–65 (2023). SPIE

    work page 2023

  8. [8]

    In: CLIP 2013, Held in Conjunction with MICCAI 2013, pp

    S´ anchez, C., Bernal, J., Gil, D., S´ anchez, F.J.: On-line lumen centre detection in gastrointestinal and respiratory endoscopy. In: CLIP 2013, Held in Conjunction with MICCAI 2013, pp. 31–38 (2014). Springer

    work page 2014

Show all 24 references
  1. [9]

    EPFL Technical Report 149300 (June 2010)

    Achanta, R., Shaji, A., Smith, K., Lucchi, A., Fua, P., S¨ usstrunk, S.: Slic superpixels. EPFL Technical Report 149300 (June 2010)

    2010

  2. [10]

    IEEE/CVF Int

    Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., Lo, W.-Y., Doll´ ar, P., Girshick, R.: Segment anything. IEEE/CVF Int. Conf. on Computer Vision, 4015–4026 (2023)

    2023

  3. [11]

    The Laryngoscope 119(7), 1318–1324 (2009)

    Murgu, S., Colt, H.G.: Morphometric bronchoscopy in adults with central airway obstruction: case illustrations and review of the literature. The Laryngoscope 119(7), 1318–1324 (2009)

    2009

  4. [12]

    International journal of pediatric otorhinolaryngology 116, 65–69 (2019) 11

    Francom, C.R., Best, C.A., Eaton, R.G., Pepper, V., Onwuka, A.J., Breuer, C.K., Lind, M.N.M., Grischkan, J.M., Chiang, T.: Clinical validation and repro- ducibility of endoscopic airway measurement in pediatric aerodigestive evaluation. International journal of pediatric otorh...

    2019

  5. [13]

    International journal of computer assisted radiology and surgery 12, 1089–1099 (2017)

    Visentini-Scarzanella, M., Sugiura, T., Kaneko, T., Koto, S.: Deep monocular 3d reconstruction for assisted navigation in bronchoscopy. International journal of computer assisted radiology and surgery 12, 1089–1099 (2017)

    2017

  6. [14]

    Scientific Data 10(1), 671 (2023)

    Azagra, P., Sostres, C., Ferr´ andez,´A., Riazuelo, L., Tomasini, C., Barbed, O.L., Morlana, J., Recasens, D., Batlle, V.M., G´ omez-Rodr´ ıguez, J.J., Elvira, R., L´ opez, J., Oriol, C., Civera, J., Tard´ os, J., Murillo, A.C., Lanas, ´A., Montiel, J.M.M.: Endomapper dataset ...

    2023

  7. [15]

    In: 8th ACM on Multimedia Systems Conference, pp

    Pogorelov, K., Randel, K.R., Griwodz, C., Eskeland, S.L., Lange, T., Johansen, D., Spampinato, C., Dang-Nguyen, D.-T., Lux, M., Schmidt, P.T., Riegler, M., Halvorsen, P.: Kvasir: A multi-class image dataset for computer aided gastroin- testinal disease detection. In: 8th ACM o...

    2017

  8. [16]

    In: MICCAI, pp

    Batlle, V.M., Montiel, J.M.M., Fua, P., Tard´ os, J.D.: Lightneus: Neural surface reconstruction in endoscopy using illumination decline. In: MICCAI, pp. 502–512. Springer, Cham (2023)

    2023

  9. [17]

    In: IEEE/RSJ Int

    Batlle, V.M., Montiel, J.M.M., Tard´ os, J.D.: Photometric single-view dense 3D reconstruction in endoscopy. In: IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), pp. 4904–4910 (2022). https://doi.org/10.1109/IROS47612. 2022.9981742

    2022

  10. [18]

    Modrzejewski, R., Collins, T., Hostettler, A., Marescaux, J., Bartoli, A.: Light modelling and calibration in laparoscopy. Int. J. Computer Assisted Radiology and Surgery 15(5), 859–866 (2020)

    2020

  11. [19]

    GitHub (2020)

    Deshpande, A.M.: Multi-object trackers in Python. GitHub (2020)

    2020

  12. [20]

    In: CVPR (2024)

    Yang, L., Kang, B., Huang, Z., Xu, X., Feng, J., Zhao, H.: Depth anything: Unleashing the power of large-scale unlabeled data. In: CVPR (2024)

    2024

  13. [21]

    In: ECCV (2016)

    Sch¨ onberger, J.L., Zheng, E., Pollefeys, M., Frahm, J.-M.: Pixelwise view selection for unstructured multi-view stereo. In: ECCV (2016)

    2016

  14. [22]

    In: CVPR (2016)

    Sch¨ onberger, J.L., Frahm, J.-M.: Structure-from-motion revisited. In: CVPR (2016)

    2016

  15. [23]

    Annals of Otology, Rhinology & Laryngology 103(4), 319–323 (1994)

    Myer III, C.M., O’Connor, D.M., Cotton, R.T.: Proposed grading system for subglottic stenosis based on endotracheal tube sizes. Annals of Otology, Rhinology & Laryngology 103(4), 319–323 (1994)

    1994

  16. [24]

    In: MICCAI, pp

    Ronneberger, O., Fischer, P., Brox, T.: U-net: Convolutional networks for biomedical image segmentation. In: MICCAI, pp. 234–241 (2015). Springer 12

    2015

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.