Pith. sign in

REVIEW 3 major objections 1 minor 1 cited by

A YOLOv8 model trained on guns, knives and a custom set of 336 blunt-object images supports real-time threat detection in surveillance video.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.3

2026-06-28 01:59 UTC pith:7QDEB4CY

load-bearing objection This applies YOLOv8 to a merged gun/knife dataset plus a 336-image mobile-camera blunt-object set but gives no metrics to support the deployment claims. the 3 major comments →

arxiv 2606.05708 v1 pith:7QDEB4CY submitted 2026-06-04 cs.CV

Real-Time Threat Detection from Surveillance Cameras using Machine Learning

Gajendra Mandal , J. P. Patra , Priyansh Mahant This is my paper
classification cs.CV
keywords object detectionYOLOv8surveillancethreat detectiongun detectionknife detectionblunt objectsreal-time
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The paper builds an automated system to spot guns, knives and blunt objects such as rods or sticks inside live camera feeds. It gathers 336 new labeled images of blunt objects with a mobile camera and adds them to an existing public collection of 7,623 gun and knife pictures. The combined set of 7,959 images trains a YOLOv8 detector whose recall and precision on the new class keep rising with longer training runs and show no overfitting. This setup targets the limits of human watchers who tire and miss events across many cameras. The authors position the result as ready for use in places like campuses, streets and transport hubs because it runs fast enough for continuous video while maintaining usable detection rates.

Core claim

The paper claims that training a YOLOv8 object detector on the merged dataset of public gun and knife images plus the new 336-image blunt-object collection produces a system that balances accuracy and speed for real-time surveillance. Extended training improves metrics on the blunt-object class without signs of overfitting. The resulting framework is presented as deployable in real-world settings such as campuses, public spaces and transportation areas.

What carries the argument

YOLOv8 object detection model trained on the consolidated 7,959-image dataset covering guns, knives and blunt objects.

Load-bearing premise

The 336 custom blunt-object images, once merged with the public gun and knife set, form a training collection that matches the variety and camera conditions of actual Indian surveillance footage closely enough for the model to generalize.

What would settle it

A separate test set of real Indian surveillance videos on which the model shows a clear drop in accuracy or frame rate below usable levels would falsify the suitability claim.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Longer training duration raises recall and average precision for the blunt object class.
  • No overfitting appears even after extended training on the merged dataset.
  • The trained model supports real-time performance needed for live surveillance feeds.
  • Detection covers the three threat classes relevant to violent activity in the target environments.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same custom-data method could fill gaps for other locally common objects not covered by public weapon sets.
  • Running the detector on low-power edge hardware would test whether it fits existing camera installations without extra servers.
  • Connecting the output directly to alert systems could shorten the time between detection and human response.
  • The gains on the blunt-object class show that public datasets alone leave out important regional variations in threat appearance.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 1 minor

Summary. The manuscript proposes a YOLOv8-based real-time object detection framework for identifying guns, knives, and blunt objects (iron/wooden/plastic rods) in surveillance videos, targeting Indian scenarios. It describes constructing a custom dataset of 336 mobile-camera images of blunt objects, merging it with a public 7,623-image gun/knife set to yield 7,959 total images, training YOLOv8 on the combined data, and asserts that longer training improves recall and average precision for the blunt class without overfitting. The central claim is that the resulting framework achieves an effective accuracy-efficiency balance suitable for deployment in real-world settings such as campuses, public spaces, and transportation areas.

Significance. If the empirical claims were supported by quantitative metrics, validation protocols, and target-domain testing, the work would offer modest practical value by extending standard weapon-detection datasets with region-specific blunt objects. The dataset-merging strategy is a simple, reproducible step that could be useful for practitioners. However, the manuscript provides no such evidence, so its significance remains limited to an initial exploration rather than a validated contribution.

major comments (3)
  1. [Abstract] Abstract: The claim that 'increasing the training duration significantly improves recall and average precision for the blunt object class without signs of overfitting' is unsupported by any reported metrics, training curves, validation splits, or error analysis. This directly undermines the central performance and deployment claims.
  2. [Abstract] Dataset construction (Abstract): The 336-image custom blunt-object set constitutes only ~4% of the merged 7,959-image collection, producing severe class imbalance (~48% gun, ~48% knife). Collection via mobile camera further introduces unaddressed domain shift in viewpoint, resolution, lighting, and motion blur relative to fixed CCTV; no target-domain evaluation on Indian surveillance footage is described, rendering the generalization claim to 'real-world surveillance environments' unsupported.
  3. [Abstract] Experimental evaluation (Abstract): No baseline comparisons, ablation studies on training duration, or quantitative results (e.g., per-class mAP, precision-recall values) are supplied to substantiate the accuracy-efficiency balance or the absence of overfitting to mobile-camera cues.
minor comments (1)
  1. [Abstract] The abstract refers to 'region-specific blunt objects commonly involved in violent activities in Indian surveillance scenarios' without citing supporting statistics or references.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the detailed and constructive comments. We agree that several claims in the abstract require supporting quantitative evidence and will revise the manuscript accordingly to strengthen the presentation of results, limitations, and evaluation protocols.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The claim that 'increasing the training duration significantly improves recall and average precision for the blunt object class without signs of overfitting' is unsupported by any reported metrics, training curves, validation splits, or error analysis. This directly undermines the central performance and deployment claims.

    Authors: We acknowledge that the abstract statement on training duration effects is not backed by reported metrics or curves in the current version. In the revision we will add per-epoch validation curves, per-class precision-recall values, and an explicit error analysis section demonstrating the observed improvements and lack of overfitting indicators (e.g., validation loss plateau). revision: yes

  2. Referee: [Abstract] Dataset construction (Abstract): The 336-image custom blunt-object set constitutes only ~4% of the merged 7,959-image collection, producing severe class imbalance (~48% gun, ~48% knife). Collection via mobile camera further introduces unaddressed domain shift in viewpoint, resolution, lighting, and motion blur relative to fixed CCTV; no target-domain evaluation on Indian surveillance footage is described, rendering the generalization claim to 'real-world surveillance environments' unsupported.

    Authors: The class distribution is indeed imbalanced; we will report the exact counts and add a discussion of class-weighting or oversampling strategies used during training. On domain shift, we agree the mobile-camera collection differs from fixed CCTV and will revise the generalization language to note this limitation explicitly while clarifying that the public gun/knife images already span varied sources. No target-domain Indian CCTV test set was collected, so we will remove or qualify the deployment claim and list target-domain evaluation as future work. revision: partial

  3. Referee: [Abstract] Experimental evaluation (Abstract): No baseline comparisons, ablation studies on training duration, or quantitative results (e.g., per-class mAP, precision-recall values) are supplied to substantiate the accuracy-efficiency balance or the absence of overfitting to mobile-camera cues.

    Authors: We will expand the experimental section with (i) quantitative tables reporting per-class mAP@0.5, precision, recall, and inference speed on the validation split, (ii) an ablation table comparing different training durations, and (iii) baseline comparisons against YOLOv5 and Faster R-CNN under identical data splits. These additions will directly support the accuracy-efficiency claims. revision: yes

Circularity Check

0 steps flagged

No circularity: purely empirical ML training with no derivations

full rationale

The paper describes dataset curation (336 custom blunt-object images merged with 7,623 public gun/knife images), YOLOv8 training, and empirical reporting of metrics such as recall and average precision after varying training duration. No equations, first-principles derivations, parameter-fitting steps presented as predictions, or self-citation chains appear in the text. All claims reduce to standard supervised learning outcomes on the provided labeled data rather than any self-referential construction. This is a conventional empirical ML report whose validity rests on external benchmarks (public datasets and standard YOLOv8 training) rather than internal reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review; the central claim rests on unstated assumptions that the custom dataset distribution matches deployment conditions and that standard YOLOv8 training generalizes from the merged collection.

axioms (1)
  • domain assumption YOLOv8 object detection can be trained on a mixed public-plus-custom image set to produce real-time inference suitable for surveillance video.
    Invoked implicitly by the choice of model and the claim of real-time performance.

pith-pipeline@v0.9.1-grok · 5762 in / 1316 out tokens · 25814 ms · 2026-06-28T01:59:45.454698+00:00 · methodology

0 comments
read the original abstract

Ensuring public safety in densely populated urban environments remains a critical challenge, necessitating the deployment of intelligent and automated video surveillance systems. Traditional surveillance approaches rely heavily on manual monitoring, which is inefficient and susceptible to human fatigue, delayed response, and observational errors. To overcome these limitations, this work presents a real-time object detection-based surveillance framework. The proposed system focuses on detecting guns, knives, and region-specific blunt objects commonly involved in violent activities in Indian surveillance scenarios. A key contribution of this work is the use of a custom-created dataset collected using a mobile camera, consisting of 336 labeled images of blunt objects such as iron rods, wooden sticks, and plastic rods. This dataset is combined with a publicly available dataset of 7,623 images of guns and knives, forming a consolidated dataset of 7,959 images across three classes: gun, knife, and blunt object. The combined dataset is used to train a YOLOv8-based object detection model for real-time performance. Experimental evaluation shows that increasing the training duration significantly improves recall and average precision for the blunt object class without signs of overfitting. Overall, the proposed framework achieves an effective balance between accuracy and efficiency, making it suitable for deployment in real-world surveillance environments such as campuses, public spaces, and transportation areas.

Figures

Figures reproduced from arXiv: 2606.05708 by Gajendra Mandal, J. P. Patra, Priyansh Mahant.

Figure 1
Figure 1. Figure 1: Workflow of the proposed methodology B. Data Collection One major limitation found in the existing datasets related to weapon detection is that the Indian public environment is poorly represented. Publicly available weapon detection datasets comprise non-Indian public scenes and mainly focus on firearms and knives. In an effort to overcome this issue, this study presents a customized dataset consisting of … view at source ↗
Figure 2
Figure 2. Figure 2: Class-wise mAP@0.5 comparison between 50 and 100 [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Epoch-wise comparison of overall mAP@0.5 showing [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Precision-Recall curves for object detection classes [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 6
Figure 6. Figure 6: Real-time testing of the proposed surveillance pipeline [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Multi-Task Deep Learning Framework for Real-Time Intelligent Video Surveillance with Temporal Event Validation

    cs.CV 2026-07 conditional novelty 4.0

    Temporal multi-frame validation around concurrent CV detectors cuts fire false alarms from 52% to 4% and lifts video plate exact-match from 66.7% to 81.8% at real-time latency on one commodity GPU.

Reference graph

Works this paper leans on

12 extracted references · 7 canonical work pages · cited by 1 Pith paper

  1. [1]

    In: 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS)

    R. Varghese and S. M., “YOLOv8: A novel object de- tection algorithm with enhanced performance and ro- bustness,” inProc. International Conference on Ad- vances in Data Engineering and Intelligent Computing Systems (ADICS), Chennai, India, 2024, pp. 1–6, doi: 10.1109/ADICS58448.2024.10533619

    work page doi:10.1109/adics58448.2024.10533619 2024

  2. [2]

    Understand- ing of a convolutional neural network,

    S. Albawi, T. A. Mohammed, and S. Al-Zawi, “Understand- ing of a convolutional neural network,” inProc. Interna- tional Conference on Engineering and Technology (ICET), Antalya, Turkey, 2017, pp. 1–6, doi: 10.1109/ICEngTech- nol.2017.8308186

    work page doi:10.1109/icengtech- 2017

  3. [3]

    Singh and C

    D. Singh and C. K. Mohan, ”Deep Spatio-Temporal Rep- resentation for Detection of Road Accidents Using Stacked Autoencoder,”in IEEE Transactions on Intelligent Trans- portation Systems, vol. 20, no. 3, pp. 879-887, March 2019, doi: 10.1109/TITS.2018.2835308

    work page doi:10.1109/tits.2018.2835308 2019

  4. [4]

    Real-Time Abnormal Object Detec- tion for Video Surveillance in Smart Cities

    Ingle, P.Y .; Kim, Y .-G. Real-Time Abnormal Object Detec- tion for Video Surveillance in Smart Cities. Sensors 2022, 22, 3862. https://doi.org/10.3390/s22103862”

    work page doi:10.3390/s22103862 2022

  5. [5]

    Indian Blunt Object Detection Dataset,

    G. Mandal, “Indian Blunt Object Detection Dataset,” Kag- gle, 2025. [Online]. Available: https://www.kaggle.com/ datasets/gajendramandalcsvtu/blunt-object-dataset

    2025

  6. [6]

    Gun and Knife Detection Dataset,

    M. Ahmed, “Gun and Knife Detection Dataset,” Roboflow Universe, 2023. [Online]. Available: https://universe. roboflow.com/mahad-ahmed/gun-and-knife-detection

    2023

  7. [7]

    Custom and Public Weapon Detection Dataset for Model Training,

    G. Mandal, “Custom and Public Weapon Detection Dataset for Model Training,” Kaggle, 2025. [Online]. Available: https://www.kaggle.com/datasets/gajendramandalcsvtu/ custum-dataset-and-public-data-for-model-training

    2025

  8. [8]

    A dataset and system for real-time gun detection in surveillance video using deep learning,

    D. Qiet al., “A dataset and system for real-time gun detection in surveillance video using deep learning,”arXiv preprint arXiv:2105.01058, 2021. [Online]. Available: https: //arxiv.org/abs/2105.01058

    work page arXiv 2021

  9. [9]

    Real-time weapon detection us- ing YOLOv8 for enhanced safety,

    A. Thakuret al., “Real-time weapon detection us- ing YOLOv8 for enhanced safety,”arXiv preprint arXiv:2410.19862, 2024. [Online]. Available: https://arxiv. org/abs/2410.19862

    work page arXiv 2024

  10. [10]

    Next-gen security: YOLOv8 for real- time weapon detection,

    D. Deshpandeet al., “Next-gen security: YOLOv8 for real- time weapon detection,” inProc. 7th Int. Conf. I-SMAC (IoT in Social, Mobile, Analytics and Cloud), 2023, pp. 1055– 1060

    2023

  11. [11]

    Literature review of deep-learning-based detection of violence in video,

    P. Negreet al., “Literature review of deep-learning-based detection of violence in video,”Sensors, vol. 24, no. 12, p. 4016, 2024. [Online]. Available: https://www.mdpi.com/ 1424-8220/24/12/4016

    2024

  12. [12]

    IEEE Transactions on Pattern Analysis and Machine Intelligence46(8), 19 5625–5644 (2024) https://doi.org/10.1109/TPAMI.2024.3369699

    J. Zhanget al., “Vision-language models for vi- sion tasks: A survey,”IEEE Trans. Pattern Anal. Mach. Intell., vol. 46, no. 8, pp. 5022–5043, 2024. doi: 10.1109/TPAMI.2024.3369699. [Online]. Available: https://arxiv.org/abs/2304.00685

    work page doi:10.1109/tpami.2024.3369699 2024