Object Detection in 20 Years: A Survey
Open accessThe DOI matches its Crossref record; 309 of the 309 references it cites were found in a registry or an index; 84 of the 192 sources its claims cite were checked for support.
Abstract
Object detection, as of one the most fundamental and challenging problems in computer vision, has received great attention in recent years. Over the past two decades, we have seen a rapid technological evolution of object detection and its profound impact on the entire computer vision field. If we consider today’s object detection technique as a revolution driven by deep learning, then, back in the 1990s, we would see the ingenious thinking and long-term perspective design of early computer vision. This article extensively reviews this fast-moving research field in the light of technical evolution, spanning over a quarter-century’s time (from the 1990s to 2022). A number of topics have been covered in this article, including the milestone detectors in history, detection datasets, metrics, fundamental building blocks of the detection system, speedup techniques, and recent state-of-the-art detection methods.
Overview
CoChat Summary
TL;DR
This survey reviews two decades of object detection research (1990s–2022), tracing the transition from handcrafted-feature detectors to deep-learning-based two-stage, one-stage, and transformer detectors. It organizes the field around milestone detectors, datasets and metrics, key building blocks (multi-scale detection, context, hard negative mining, loss functions, NMS), speed-up techniques, and recent advances.
Key findings
- The field splits into a "traditional object detection period (before 2014)" and a "deep learning based detection period (after 2014)"; RCNN raised VOC07 mAP from 33.7% (DPM-v5) to 58.5%, SPPNet reached VOC07 mAP=59.2% at >20× faster than R-CNN, Fast RCNN reached 70.0% mAP at >200× faster than R-CNN, and Faster RCNN reached COCO mAP@.5=42.7%, VOC07 mAP=73.2% at 17fps.
- One-stage detectors traded some accuracy for speed: YOLO ran at 155fps with VOC07 mAP=52.7% (enhanced version 45fps, 63.4%), SSD reached COCO mAP@.5=46.5% at 59fps, and RetinaNet's focal loss brought one-stage accuracy to COCO mAP@.5=59.1%.
- Keypoint- and transformer-based detectors pushed accuracy further: CornerNet COCO mAP@.5=57.8%, CenterNet 61.1%, and Deformable DETR 71.9% on MS-COCO; the paper notes the top-10 COCO detection methods are all transformer-based.
- Datasets scaled dramatically: VOC07 has 5k train images/12k objects, VOC12 11k/27k, MS-COCO-17 164k images and 897k annotated objects over 80 categories, and Open Images 1,910k images with 15,440k boxes over 600 categories. Evaluation moved from miss-rate vs. FPPW to FPPI, then to AP/mAP, with MS-COCO averaging over IoU thresholds 0.5–0.95.
Methods
- Design: a narrative literature survey spanning roughly a quarter-century (1990s to 2022), structured around technical evolution rather than a systematic/quantitative review protocol.
- Coverage: milestone detectors (VJ, HOG, DPM, RCNN, SPPNet, Fast/Faster RCNN, FPN, YOLO, SSD, RetinaNet, CornerNet, CenterNet, DETR), datasets and metrics, technical building blocks, speed-up techniques, and recent advances (rotation/scale robustness, backbones, localization, segmentation loss, adversarial training, weakly supervised detection, domain adaptation).
- Analysis: qualitative synthesis of reported detector accuracies and speeds, with figures/tables summarizing dataset statistics and accuracy trends; no new experiments are conducted.
Limitations
- The paper states no explicit limitations section; it frames its scope as presenting the evolution of technologies "while neglecting their technical specifics," and notes that the evolution order shown in figures is primarily to assist reader understanding and "may have temporal overlap."
- Evident from the text: as a survey it reports no original experiments or controlled benchmarks, so the cited mAP/speed figures come from the original papers and are not independently verified; the coverage is also selective (milestone detectors and selected recent advances) rather than exhaustive.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
55 Valid · 18 Not valid · 32 Not checked · 50 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“In the deep learning era, due to the increase of computing power, bootstrap was shortly discarded in object detection during 2014-2016 [16-20].”
What the sources say
- Not checked — no full text of the sourcecites [16] matched to “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation” (2014) source
- Disputedcites [17] matched to “Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition” (2014) sourceExperimental — accuracy not yet measured
Section 4.1 (page unavailable) says the authors used standard hard negative mining, iterated once, to train their object-detection SVMs. That is contrary to citing this 2014 detector as evidence that bootstrapping had been discarded. The paper does not establish a 2014–2016 trend or attribute one to increased computing power. (LLM, conf=0.78)
Checked by a reasoning model — may be wrong.
- Not checked — no full text of the sourcecites [18] matched to “Fast R-CNN” (2015) source
- Partly supportedcites [19] matched to “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks” (2017) sourceExperimental — accuracy not yet measured
Section 4.2 (p. 10) says Fast R-CNN abandoned the SVM step that, in SPPnet, used hard-negative mining. This supports a narrower change in one detection system, but not the claim that bootstrapping was discarded across object detection during 2014–2016 or that increased computing power caused the change. (LLM, conf=0.83)
- Could not verifycites [20] matched to “You Only Look Once: Unified, Real-Time Object Detection” (2016) sourceExperimental — accuracy not yet measured
The Introduction (p. 1) contrasts YOLO with earlier detection methods, and Training (p. 3) describes how YOLO is trained. Neither passage addresses whether bootstrapping was discarded in object detection during 2014–2016 or attributes such a change to increased computing power; searches for bootstrapping and hard-negative mining found no passage that does. (LLM, conf=0.72)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Faster RCNN is the first near-realtime deep learning detector (COCO mAP@.5=42.7%, VOC07 mAP=73.2%, 17fps with ZF-Net [48]).”
What the sources say
- Partly supportedcites [19] matched to “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks” (2017) sourceExperimental — accuracy not yet measured
Section 4.2 reports 42.7% COCO mAP@0.5; Section 4.1 reports 73.2% VOC 2007 mAP and 17 fps with ZF-Net. These are results from different model configurations: the 73.2% VOC result uses VGG-16, not ZF-Net. The paper describes Faster R-CNN as a fast detector but does not establish that it was the first near-real-time deep-learning detector. (LLM, conf=0.88)
- Partly supportedcites [47] matched to “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks” (2017) sourceExperimental — accuracy not yet measured
Section 4.2 reports 42.7% COCO mAP@0.5, and Section 4.1 reports 73.2% VOC 2007 mAP and 17 fps with ZF-Net. The accuracy figures are from VGG-16 experiments, not the 17-fps ZF-Net configuration. The paper describes Faster R-CNN as a fast detector but does not establish that it was the first near-real-time deep-learning detector. (LLM, conf=0.88)
- Partly supportedcites [48] matched to “Visualizing and understanding convolutional networks” (2014) sourceExperimental — accuracy not yet measured
Section 3 (p. 4) describes the authors’ convolutional-network architecture, supporting the citation as an identification of the ZF-Net backbone. The paper does not report Faster R-CNN’s near-real-time status, COCO or VOC07 mAP, or 17-fps speed, so it does not support those performance claims. (LLM, conf=0.88)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“In the case of sliding window based detectors, the imbalance between backgrounds and objects could be as extreme as 10 7 :1 [71].”
What the sources say
- Partly supportedcites [71] matched to “What Makes for Effective Detection Proposals?” (2016) sourceExperimental — accuracy not yet measured
The Introduction (p. 1) says sliding-window detection can search 10^6–10^7 windows per image. It does not say that 10^7 of those windows are backgrounds for each object, so it does not establish the claimed 10^7:1 imbalance. (LLM, conf=0.88)
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“It achieves state-of-the-art performance on MSCOCO dataset (COCO mAP@.5=71.9%).”
What the sources say
- Supportedcites [43] Zhu, 2020, Deformable DETR: Deformable transformers for end-to-end object detection, arXiv:2010.04159 source
Section 5.3 (pp. 8–9) compares Deformable DETR with state-of-the-art methods on COCO. Its table reports an AP50 of 71.9 for the ResNeXt-101 + DCN configuration with test-time augmentation; the accompanying text reports 52.3 overall AP for that configuration. (LLM, conf=0.91)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Using FPN in a basic Faster R-CNN system, it achieves state-of-the-art single model detection results on the COCO dataset without bells and whistles (COCO mAP@.5=59.1%).”
What the sources say
- Supportedcites [24] matched to “Feature Pyramid Networks for Object Detection” (2017) source
The Introduction describes a state-of-the-art single-model COCO result using FPN with a basic Faster R-CNN detector, without bells and whistles. Section 5.2.3 reports AP@0.5 of 59.1 on COCO test-dev and says the single-model entry surpassed competition winners. Page numbers are not provided in the copy. (LLM, conf=0.94)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
50 more claims · 21 with their cited sources checked
Create a free account to see all 155 claims and to verify the 3 references still open.
- It may state two separate claims as one. 40
- Probably not a substantial claim. 83
- The quote only partly states this claim. 1
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“From navigating through occlusions to efficiently managing scale variations and fine-grained object detection scenarios [9], YOLO faces a myriad of obstacles.”
From YOLOv1 to v8: Unveiling Each Variant–A Comprehensive Review of YOLO (2024) · cites this paper as [9]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“The speed-up techniques in object detection can be divided into three levels of groups: speed up of 'detection pipeline', 'detector backbone', and 'numerical computation'. , as shown in Fig. 9. Refer to [123] for a more detailed version.”
From Object Detection in 20 Years: A Survey (2023) · cites this paper as [123]
The cited source says what the claim attributes to it.
In the citing paper
“The few surveys that mention both object detection and model compression [11] focus on the acceleration of”
From A survey of model compression strategies for object detection (2023) · cites this paper as [11]
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“In addition to that, most of the reviews in related fields review the object detection model and model compression technology separately, and there is a lack of comprehensive reviews for both.”
From A survey of model compression strategies for object detection (2023) · cites this paper as [11]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ II. OBJECT DETECTION IN 20 YEARS
“Fig. 2: A road map of object detection. Milestone detectors in this figure: VJ Det. [10, 11], HOG Det. [12], DPM [1315], RCNN [16], SPPNet [17], Fast RCNN [18], Faster RCNN [19], YOLO [20-22], SSD [23], FPN [24], Retina-Net [25], CornerNet [26], CenterNet [27], DETR [28].”
§ Object detection accuracy improvements
“CenterNet: X. Zhou et al proposed CenterNet [40] in 2019.”
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 4×
“As the accuracy of a detector depends heavily on its feature extraction networks, in this paper, we refer to the backbone networks, e.g. the ResNet and VGG, as the “engine” of a detector.”
InfluentialImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 5×
“In 2012, the world saw the rebirth of convolutional neural networks (CNNs) [35].”
InfluentialLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472 · cited 10×
“The main idea of multi-reference detection [19, 22, 23, 41, 47, 51] is to first define a set of references (a.”
InfluentialFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“It contains a detection challenge using ImageNet images [61].”
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 3×
“Some well-known detection models of the early time: (a) Eigenfaces [95], (b) Shared weight networks [96], (c) Space displacement networks (Lenet-5) [97], (d) Haar wavelets of VJ detector [10].”
BackgroundMethodDistinctive Image Features from Scale-Invariant Keypoints
Lowe, D. G. (2004) · International Journal of Computer Vision · doi:10.1023/b:visi.0000029664.99615.94 · Cited by 55,589 · cited 1×
“HOG can be considered as an important improvement of the scale-invariant feature transform [33, 34] and shape contexts [35] of its time.”
BackgroundYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 9×
“L2 loss is prevalent in early research [16], [20], [51], but it is highly affected by outliers and prone to gradient explosion.”
InfluentialDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Research works printed but not linked: 9
- 10.5555/3045118.3045336 · doi:10.5555/3045118.3045336
- Girshick, 2023, Discriminatively Trained Deformable Part Models, Release 5
- 10.1109/TPAMI.2016.2577031 · DOI · Paper page
- Krasin, 2017, OpenImages: A Public Dataset for Large-Scale Multi-Label and Multi-Class Image Classification
- Alexe, What is an object, 73
- Müller, When does label smoothing help, 32, 1
- 10.1109/CVPR.2016.308 · DOI · Paper page
- 10.5555/2969033.2969125 · doi:10.5555/2969033.2969125
- 10.1109/ICCV.2017.244 · DOI · Paper page
Cited by
Items 1,201–1,300 of 3,144
Wu, Y., & Chen, J. (2023) · International Journal on Semantic Web and Information Systems · doi:10.4018/ijswis.330015 · Cited by 7
Single Plant Detection and Isolation on High Resolution UAV Data
Spielbauer, N., Blumenthal, D., Puck, L., Heppner, G., Strathmann, M., Bauer, C., Roth, M., Mink, R., Linn, A. I., Roennau, A., et al. (2023) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case56687.2023.10260423 · Cited by 5
Hussain, T. A., Memon, Z. A., Qureshi, R., et al. (2023) · Sensors · doi:10.3390/s23198106 · Cited by 4
“…activation functions, these Convolution neural networks have been widely used for image-related tasks such as classification [20], object detection [21,22], and instance segmentation [23], and video classification [24]. architectures can overcome the limitations of traditional models, like…”
BackgroundMethodTransformer-based End-to-End Object Detection in Aerial Images
Vo, N. D., Le, N., Ngo, G. H., Doan, D., Le, D., & Nguyen, K. (2023) · International Journal of Advanced Computer Science and Applications · doi:10.14569/ijacsa.2023.01410113 · Cited by 5
SMS-based Dog Detection in Residential Area using YOLOv5 and ML.Net
Galeno, P. C., Sale, D. P., & Martin, E. M. B. (2023) · IEEE Conference on Systems, Process & Control (ICSPC) · doi:10.1109/icspc59664.2023.10420223 · Cited by 4
“Issues in object detection include accurate localization, dense and obstructed detection, object rotation and scale changes, and speed-up of detection [7].”
BackgroundTriyanto, W. A., Adi, K., & Suseno, J. E. (2024) · Mathematical Modelling and Engineering Problems · doi:10.18280/mmep.110227 · Cited by 4
Detection of Defective Products in Injection Molding Process using YOLO-NAS
Asadi, M., Hashemi, S. S., & Sadeghi, M. T. (2023) · International Conference on Control, Instrumentation, and Automation · doi:10.1109/iccia61416.2023.10506371 · Cited by 4
“Object detection has a wide range of practical applications in diverse fields, including Autonomous Vehicles, Surveillance and Security, Manufacturing and Quality Control, etc. [14].”
BackgroundJin, G., Liu, Q., Cai, W., Li, M., & Lu, C. (2024) · Applied Sciences · doi:10.3390/app14104226 · Cited by 5
Santiago, J. S. B., Farias, V. A. E., Sena, L., & Gomes, J. P. P. (2024) · Learning and Nonlinear Models · doi:10.21528/lnlm-vol22-no1-art2 · Cited by 4
Computer-Vision-Oriented Adaptive Sampling in Compressive Sensing
Liu, L., Nishikawa, H., Zhou, J., Taniguchi, I., & Onoye, T. (2024) · Sensors · doi:10.3390/s24134348 · Cited by 5
“Object detection requires identifying and locating objects of interest within an image, such as detecting and locating multiple faces in a crowd [47,48].”
BackgroundFast supervoxel segmentation of connectivity median simulation based on Manhattan distance
Yuan, Y., Wang, J., Li, W., Wang, K., Rao, H., & Xu, J. (2024) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2024.104108 · Cited by 4
Wang, W., Chen, Q., Shen, Y., et al. (2024) · Sensors · doi:10.3390/s24175569 · Cited by 4
“These computer vision techniques generally can be divided into four categories, including image classification [21], object detection [22], semantic segmentation [23], and instance segmentation [24].”
BackgroundCoarse-to-Fine Target Detection for HFSWR With Spatial-Frequency Analysis and Subnet Structure
Zhang, W., Yang, Y., & Liu, T. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3453044 · Cited by 4
“Despite the effectiveness of DL methods, recent findings [27] indicate that they may not always be optimal for remote sensing. ii) These strategies depend solely on spatial information for ship target identification.”
PhosopNet: An improved grain localization and classification by image augmentation
Mookdarsanit, P., & Mookdarsanit, L. (2021) · TELKOMNIKA (Telecommunication Computing Electronics and Control) · doi:10.12928/telkomnika.v19i2.18321 · Cited by 5
Accelerated Video Annotation Driven by Deep Detector and Tracker
Price, E., & Ahmad, A. (2024) · Lecture notes in networks and systems · doi:10.1007/978-3-031-44981-9_12 · Cited by 5
“Deep convolutional networks have also revolutionized visual object detection [10,11].”
BackgroundBeratoğlu, M. S., & Töreyın, B. U. (2023) · SSRN Electronic Journal · doi:10.2139/ssrn.4629126 · Cited by 5
XU, G., Huang, W., & Jia, W. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4893335 · Cited by 7
Evaluating Methods for Violence Classification and Firearm Detection in Indoor CCTV Environment
da Silva, A. V. B., & Pereira, L. F. A. (2024) · Journal of the Brazilian Computer Society · doi:10.5753/jbcs.2024.3282 · Cited by 3
“The utilization of learned (LN) features has stood out in various areas such as image classification [Lu and Weng, 2007], object detection [Zou et al ., 2019], image segmentation [Garcia-Garcia et al ., 2017].”
BackgroundSajja, B., & Kim, S. H. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10826076 · Cited by 3
Biosafety Management System for Pig Farms Based on Deep Learning and Internet of Things Technology
Liang, K., & Wang, M. (2024) · China Automation Congress (CAC) · doi:10.1109/cac63892.2024.10864547 · Cited by 4
“It is mainly divided into two-stage detection algorithm and one-stage detection algorithm[3].”
Tong, Y., Liu, J., Cao, G., Li, L., & Wang, Y. (2025) · European Journal of Remote Sensing · doi:10.1080/22797254.2025.2469863 · Cited by 5
“Unlike traditional object detection methods that heavily rely on manually crafted object features (Park et al., 2015), deep learning-based object detectors provide more efficient feature extraction capabilities and superior object detection performance (Zou et al., 2023).”
A Multimodal Scale Normalization Framework for Vision-Radar Small UAV Positioning
Wan, Y., Ji, J., Xie, W., Wu, G., Zhou, F., & Wu, Q. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3549620 · Cited by 3
“In general, constructing a feature pyramid to obtain feature from different stages of the backbone network can mitigate this issue but not fundamentally [36].”
CrowdRadar: a mobile crowdsensing framework for urban traffic green travel safety risk assessment
Wang, Y., Tang, Q., Wei, W., Yang, C., Yang, D., Wang, C., Xu, L. J., & Chen, L. (2025) · Frontiers in Big Data · doi:10.3389/fdata.2025.1440816 · Cited by 4
A High‐Accuracy YOLOv8‐ResAttNet Framework for Maritime Vessel Detection Using Residual Attention
Liu, P., Sun, M., Han, X., Liu, S., Chen, Y., & Zhang, H. (2025) · IET Image Processing · doi:10.1049/ipr2.70085 · Cited by 6
YOLO-G3CF: Gaussian Contrastive Cross-Channel Fusion for Multimodal Object Detection
Belmouhcine, A., Pham, M., & Lefèvre, S. (2025) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2025.3564181 · Cited by 10
“O BJECT detection continues to be a central focus in both computer vision [1] and remote sensing [2].”
Jafari, A. A., Agarwal, A., Ozcınar, C., & Anbarjafari, G. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-04243-z · Cited by 3
An X‐Ray Contraband Detection Method Based on Improved YOLOv8
Chen, J., Hao, J., & Liu, X. (2025) · IET Image Processing · doi:10.1049/ipr2.70135 · Cited by 7
UTNet: event-RGB multimodal fusion model for underwater transparent organism detection
Guo, F., Ren, P., & Luo, C. L. (2025) · Intelligent Marine Technology and Systems · doi:10.1007/s44295-025-00065-4 · Cited by 6
“They record the entire scene at a fixed frame rate and output the data as a sequence of discrete frames (Zou et al. 2023).”
Sowmitha, V., Hemalatha, S., Indumathi, M., et al. (2025) · doi:10.1109/icima64861.2025.11074015 · Cited by 3
“This technology lessens reliance on manual inspection for auditing and supervision, automating the entire procedure, leaving the parking ambiance secure and sound further[1].”
Oriol, P., Baños-Castelló, P., Padró, E. M., Francescangeli, M., Toma, D. M., Widmer, M. C., Farran, J. D. P., & Fernández, J. D. R. (2025) · Artificial intelligence · doi:10.5772/intechopen.1011280 · Cited by 4
Benelmostafa, B.-E., Aitelhaj, R., & Medromi, H. (2025) · IEEE Access · doi:10.1109/access.2025.3593201 · Cited by 7
“Attention mechanisms significantly enhance deep learning performance by enabling models to selectively focus on crucial input information, thereby improving accuracy in image processing tasks such as object detection [36].”
A Context-Aware Doorway Alignment and Depth Estimation Algorithm for Assistive Wheelchairs
Tennekoon, S., Wedasingha, N., Welhenge, A. M., Abhayasinghe, N., & Murray, I. D. (2025) · Computers · doi:10.3390/computers14070284 · Cited by 4
A Comprehensive Hybrid Approach for Indoor Scene Recognition Combining CNNs and Text-Based Features
Uckan, T., Aslan, C., & Hark, C. (2025) · Sensors · doi:10.3390/s25175350 · Cited by 5
“Moreover, object recognition serves as a foundational step for various downstream tasks, including object detection and tracking, image captioning, and semantic segmentation [24].”
He, Y., Su, H., Yu, H., Yang, C., Wei, S., Wang, C., & Liu, S. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11127317 · Cited by 8
“Significant advances in this field have been driven by the advent of convolutional neural networks (CNNs) [4].”
Liu, S. H., Li, K., Ding, Y., et al. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.110944 · Cited by 3
Monzón-Verona, J. M., García–Alonso, S., & Santana-Martín, F. J. (2025) · Electronics · doi:10.3390/electronics14193916 · Cited by 3
Tiny object detection based on dynamic scale-awareness label assignment and contextual enhancement
Zhang, T., Zhang, X., Hua, C., Wang, G., Han, X., & Jiao, L. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.112449 · Cited by 9
Message-passing framework for multi-camera worker tracking in construction
Khan, N., Kim, D., Kim, M., Kim, D., & Lee, D. (2025) · Automation in Construction · doi:10.1016/j.autcon.2025.106610 · Cited by 9
Gillani, H. H., Qureshi, M. A., Beghdadi, A., Cheikh, F. A., & Ullah, M. (2025) · ACM Computing Surveys · doi:10.1145/3773023 · Cited by 9
“Unlike semantic classification problems with a large body of research [199], distortion classification has seen limited focus, often being an auxiliary step in IQA/VQA tasks [40, 44, 114, 148, 188–190].”
Miao, W., Shen, J., Xu, H., Kärkkäinen, T., Xu, Q., Xu, Y., & Cong, F. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754854 · Cited by 9
“One of the fundamental tasks is object detection [50].”
Deep learning-based forest fire detection using an improved SSD algorithm with CBAM
Zhang, D., Zhang, Y., Dong, L., Ruan, S., & Liu, Z. (2025) · PLoS ONE · doi:10.1371/journal.pone.0333574 · Cited by 9
“With the increasing adoption of computer vision technologies, significant progress has been made in various domains such as surveillance systems, medical diagnostics, and image retrieval [2].”
CHCG-YOLO: a lightweight model for fruit recognition
An, X., Sun, L., Zhao, Y., Wang, H., Zhang, M., Li, K., & Xie, R. (2025) · Journal of Food Measurement & Characterization · doi:10.1007/s11694-025-03775-9 · Cited by 4
ControlLoc: Physical-World Hijacking Attack on Camera-based Perception in Autonomous Driving
Ma, C., Wang, N., Zhao, Z., Wang, Q., Chen, Q. A., & Shen, C. Q. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3744842 · Cited by 18
“The AD perception algorithm employs an object detector [77] to generate a bounding box (BBOX) and classify the object.”
Multiple Large AI Models’ Consensus for Object Detection—A Survey
Iwanowski, M., & Gahbler, M. (2025) · Applied Sciences · doi:10.3390/app152412961 · Cited by 9
Prete, R. D., Mazzeo, A., Graziano, M. D., & Renga, A. (2025) · IEEE Journal of Oceanic Engineering · doi:10.1109/joe.2025.3619284 · Cited by 3
“FCOS adopted a pixelwise bounding box regression approach, while FSAF dynamically selected features to optimize performance [1].”
Ouyang, M., Zeng, B., & Huang, G. (2025) · Remote Sensing Applications Society and Environment · doi:10.1016/j.rsase.2025.101827 · Cited by 3
An Event-Based Fast Intensity Reconstruction Scheme for UAV Real-Time Perception
Dong, X., Zhang, Y., Cui, Y., et al. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3607233 · Cited by 3
“C ONVENTIONAL RGB cameras are widely used in uncrewed aerial vehicle (UAV) onboard object detection [1], simultaneous localization and mapping (SLAM) [2], and other visual perception domains [3], [4].”
Adaptive Grid Selection Training Strategy for Tiny Object Detection
Yim, S., Cho, M., Kang, D., & Lee, S. (2025) · IEEE Access · doi:10.1109/access.2025.3613741 · Cited by 4
“Region classification, often using Support Vector Machines (SVMs), assigned categorical labels to each region [1], [17], [18], [19].”
Zhai, Y. J., Zhang, Z., Xie, S., Tong, C., Luo, X., Li, X., Wang, L., & Zhao, Y. (2026) · Electronics · doi:10.3390/electronics15010211 · Cited by 3
Tong, Y., Yuan, J., & Hu, C. (2026) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2026.3651729 · Cited by 4
“First, the task-agnostic nature of RES and the architectural flexibility of CLC allow them to be adapted to other visual tasks, such as object detection [45], [46] and semantic segmentation [47], [48].”
A Novel Vision-Based Approach to Test Sequence Generation for Mobile GUI Testing
Cui, C., Huang, Y., Huang, R., Zhou, L., Wang, R., & Li, T. (2026) · IEEE Transactions on Reliability · doi:10.1109/tr.2025.3648728 · Cited by 8
“2) Efficiency metric (FPS): FPS is used to evaluate the running efficiency of the detection model, which is critical for VTSG’s deployment on resource-constrained mobile devices [47], [48].”
InfluentialPDDNet: An end-to-end object detection framework for real-world plant leaf disease diagnosis
Yang, F., Ma, W., Chen, Q., Sun, Y., & Han, Y. (2026) · Expert Systems with Applications · doi:10.1016/j.eswa.2026.131294 · Cited by 5
Zhang, J., Ji, X., Zhao, M., et al. (2026) · Journal of Applied Geophysics · doi:10.1016/j.jappgeo.2026.106128 · Cited by 3
A Review of Object Detection in Aerial Images by Deep Learning Approaches
Vijai, A., Padmavathi, S., & Venkataraman, D. (2026) · SN Computer Science · doi:10.1007/s42979-025-04685-9 · Cited by 6
“Generally, horizontal bounding box (HBB) is used to localize the objects where each box has (x,y,w,h) as parameters [8, 9], where (x,y) represent the central co-ordinates and (w,h) represent the width and height of the box respectively.”
UAV Small Target Detection Method Based on Frequency-Enhanced Multi-Scale Fusion Backbone
Zhong, Y., Zhao, D., Han, Y., & Wang, Z. (2026) · Drones · doi:10.3390/drones10020106 · Cited by 7
Quality over quantity: a data-centric survey of annotation errors in object detection datasets
Hussain, A., Ullah, K., Afaq, M., Munsif, M., Hussain, A., & Baik, S. W. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11502-z · Cited by 9
Litter detection from aerial imagery: a review of UAV-based approaches and deep learning techniques
Bartolo, M., Hili, G., Seychell, D., Montebello, M., Debono, C. J., Formosa, S., & Makantasis, K. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21440-1 · Cited by 6
“Object Occlusion : Object occlusion is another common challenge in detection tasks and can be categorized into partial occlusion and complete occlusion [26, 28, 29 ].”
Jurado, U. T., & Martínez-Hernández, U. (2026) · IEEE Sensors Letters · doi:10.1109/lsens.2026.3676656 · Cited by 3
Suo, Y., Wu, Y., Diao, W., et al. (2026) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2026.03.047 · Cited by 3
Saddi, K. C., van Emmerik, T. H. M., Miglino, D., Poggi, M., Isgrò, F., Tasseron, P., Daniele, L., & Manfreda, S. (2026) · Water Resources Research · doi:10.1029/2025wr040605 · Cited by 3
“Specifically, we focused on the data set quality, and understanding the level of information that can be extracted using machine learning (ML), rather than on the algorithm itself—whose development is already advancing rapidly (Zou et al., 2023).”
InfluentialXu, W., Liang, C., Lian, J., et al. (2026) · Solar Energy · doi:10.1016/j.solener.2026.114673 · Cited by 3
Ghifary, M., Nugroho, M. A. A., Barokah, B., Purwarianti, A., Birowo, C. T., Machdi, I., Gaol, F. L., Purnama, I. K. E., Riandi, O., Azhar, T. N., et al. (2023) · Penerbit BRIN eBooks · doi:10.55981/brin.668 · Cited by 7
Review of Object Detection Challenges in Autonomous Driving
Cao, S. (2023) · Applied and Computational Engineering · doi:10.54254/2755-2721/8/20230306 · Cited by 3
YOLO Models for Fresh Fruit Classification From Digital Videos
Xue, Y., & Yan, W. Q. (2023) · Advances in computational intelligence and robotics book series · doi:10.4018/978-1-6684-9999-3.ch017 · Cited by 5
In-Rack Test Tube Pose Estimation Using RGB-D Data
Chen, H., Wan, W., Matsushita, M., Kotaka, T., & Harada, K. (2023) · IEEE International Conference on Robotics and Biomimetics (ROBIO) · doi:10.1109/robio58561.2023.10354984 · Cited by 3
“The advent of convolutional neural networks (CNNs) changed this landscape dramatically [7].”
BackgroundFu, G., Wang, Y., Yang, J., Wang, S., & Yang, G. (2023) · Journal of Intelligent & Robotic Systems · doi:10.1007/s10846-023-02024-9 · Cited by 3
“Zou et al. [23] propose that two-stage detectors adopt a “coarse-to-fine” detection process, which provides good detection performance, while the multiple detection stages raise the complexity of its structure.”
Probabilistic Modeling and Control for Multi-UAV Search Over Uneven Terrain
Lanča, L., Jakac, K., & Ivić, S. (2025) · IEEE Transactions on Robotics · doi:10.1109/tro.2025.3645884 · Cited by 3
Water-Puddle Segmentation Using Deep Learning in Unstructured Environments
Kumar, A., & Choudhary, A. (2023) · IEEE International Conference on Service Operations and Logistics, and Informatics (SOLI) · doi:10.1109/soli60636.2023.10425657 · Cited by 3
“…current method of deep learning techniques based on CNN, like YOLO (You Only Look Once) [2], SSD (Single Shot Detector) [8], etc. Object detection [3] [11] and segmentation are the two most import methods of Computer Vision which requires large level of computation for it to work with high level…”
MethodAlgae Object Detection Algorithm Based on Improved YOLOv5
Lin, K., Tang, Y., Tang, J., et al. (2023) · International Conference on Software Engineering and Computer Science (CSECS) · doi:10.1109/csecs60003.2023.10428252 · Cited by 3
“The most direct method is to use a variety of physical and chemical methods to detect algal cells and their contents (cytoplasm, DNA, RNA, etc.) or indirectly detect algal secretions to obtain relevant data[2][3], and over the years, researchers have published many related research results[4].”
Defect detection in textile manufacturing: Improving findings using YOLOv5 versions
Seidel, R., Waiandt, C. A., & Júnior, H. S. (2023) · doi:10.5753/wvc.2023.27534 · Cited by 3
“A nova medida realiza um c´alculo da m´edia obtida considerando v´arios limiares simultaneamente, entre 0,5 e 0,95, caracterizando o mAP@0.5:0.95 [15].”
BackgroundInfluentialA lightweight algorithm for small traffic sign detection based on improved YOLOv5s
Cai, K., Yang, J., Ren, J., & Zhang, W. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03118-z · Cited by 3
“Object detection methods based on CNN can be broadly categorized into two groups: regression-based and region-proposal based [12].”
YOLOv8-lite: An Interpretable Lightweight Object Detector for Real-Time UAV Detection
Lai, H., Liu, B., Kan, H. Y., et al. (2023) · International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc59590.2023.10507293 · Cited by 3
“From [18], we summarize the differences between two-stage and one-stage object detection in Table I.”
MethodTennis Ball Recognition in Complex Scenes Based on Improved YOLOv5
Li, Q., & Zhao, Y. (2024) · International Conference on Advanced Algorithms and Control Engineering (ICAACE) · doi:10.1109/icaace61206.2024.10548503 · Cited by 3
Yang, J., Lee, E., Kwon, J., Lee, D., Kim, Y., Park, C., & Lee, D. (2024) · IEEE Access · doi:10.1109/access.2024.3423697 · Cited by 6
“Object detection, including object localization and classification, provides one of the most basic pieces of information for various tasks of computer vision [19], [20].”
BackgroundDevelopment Life Cycle for Using a DNN to Complete the Autonomous Aerial Refueling Task
Clement, D., Mottino, S., & Costello, D. H. (2024) · IEEE Access · doi:10.1109/access.2024.3427668 · Cited by 3
“Object detection neural networks are an instance of supervised learning, which uses labeled datasets to train algorithms to classify data and/or predict outcomes [18], [19], [20].”
BackgroundSpatial Conceptual Modeling: Anchoring Knowledge in the Real World
Fill, H.-G. (2024) · Springer eBooks · doi:10.1007/978-3-031-56862-6_3 · Cited by 4
On Efficient Object-Detection NAS for ADAS on Edge devices
Gupta, D., Lee, R. D., & Wynter, L. (2024) · IEEE Conference on Artificial Intelligence (CAI) · doi:10.1109/cai59869.2024.00183 · Cited by 5
“With the recent advances in deep learning, convolutional neural networks (CNN) models have emerged as achieving unprecedented improvement in accuracy for object detection tasks over traditional handcrafted approaches [1]–[3].”
BackgroundPest-YOLO: A Lightweight Pest Detection Model Based on Multi-level Feature Fusion
Zhu, X., Jia, B., Huang, B., Li, H., Liu, X., & Seah, W. K. G. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-5591-2_12 · Cited by 4
Lyu, H., Qiu, F., An, L., Stow, D. A., Lewision, R., & Bohnett, E. (2024) · Sensing and Imaging · doi:10.1007/s11220-024-00496-z · Cited by 4
“The idea behind this method is to apply a threshold value to a specific image feature, such as color, intensity, or texture, and then consider the regions where the pixels surpass the threshold [4, 36].”
Cho, H. (2024) · IEEE International Conference on Omni-layer Intelligent Systems (COINS) · doi:10.1109/coins61597.2024.10622125 · Cited by 6
“Object detection technology is primarily employed for waste recognition, with various methodologies available such as mask-RCNN, YOLO, among others [10-13].”
MethodYanfu, 李. L., Haiyue, 兰. L., Jingfan, 薛. X., Jinlin, 郭. G., Ruijie, 黄. H., & Jiangping, 朱. Z. (2024) · Chinese Journal of Lasers · doi:10.3788/cjl240474 · Cited by 3
DragFly: Joint Threat Object Detection And UAV Trajectory Planning in Hostile Environments
Thouchamongkol, A., Chen, H., Wang, Y., Dai, Q. T. D., & Nguyen, V.-L. (2024) · IEEE Vehicular Technology Conference · doi:10.1109/vtc2024-spring62846.2024.10683537 · Cited by 4
“In collaboration with Ultralytics, Deci has introduced YOLO-NAS for real-time object detection [11].”
MethodExploration of Data Augmentation Techniques for Bush Detection in Blueberry Orchards
Čuljak, B., Pajević, N., Vladan, F., Dimitrije, S., Grbović, Ž., Djuric, N., & Panić, M. N. (2024) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw63382.2024.00576 · Cited by 3
“Both classification and objectness losses are computed using binary cross-entropy, while the location loss is derived from the complete intersection-over-union (IoU) loss [38, 44].”
MethodDetection and Identification of Hazardous Hidden Objects in Images: A Comprehensive Review
Swain, S., & Devi, K. S. (2024) · Archives of Computational Methods in Engineering · doi:10.1007/s11831-024-10173-9 · Cited by 7
“It deals with the detection of semantic objects of specific types such as visual (cars, animals, persons, etc.), hidden (guns, knives, bombs, etc.), or concealed objects in images, videos, and real-time scenarios [1].”
Detection and removal of excess materials in aircraft wings using continuum robot end-effectors
Cao, X., Yu, J., Tang, S., Sui, J., & Pei, X. (2024) · Frontiers of Mechanical Engineering · doi:10.1007/s11465-024-0806-2 · Cited by 3
Swin‐YOLOX for autonomous and accurate drone visual landing
Chen, R., Xu, Y., bin Sinal, M. S., Zhong, D., Li, X., Li, B., Guo, Y., & Luo, Q. (2024) · IET Image Processing · doi:10.1049/ipr2.13282 · Cited by 4
Self-citationEfficient Deep Learning Models for Litter Detection in the Wild
Bianco, S., Gaviraghi, E., & Schettini, R. (2024) · IEEE Forum on Research and Technologies for Society and Industry Innovation (RTSI) · doi:10.1109/rtsi61910.2024.10761805 · Cited by 3
“YOLO belongs to the category of one-stage detectors [16] and spatially separates bounding boxes and associates probabilities to each of the detected object using a single pass over the input image with a Convolutional Neural Network (CNN).”
Liang, C., Wang, Z.-Z., Liu, X.-L., Zhang, P., Tian, Z.-W., & Qian, R.-L. (2024) · IEEE Access · doi:10.1109/access.2024.3512185 · Cited by 9
SkyML: A MLaaS Federation Design for Multicloud-Based Multimedia Analytics
Xie, S., Xue, Y., Zhu, Y., & Wang, Z. (2024) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2024.3521768 · Cited by 6
“Object detection [17] is a versatile vision task that aims to detect the locations and categories of all objects in an image, with applications across diverse multimedia domains [18], [19], [20], such as video analytics [21], [22] and autonomous driving [23], [24].”
Unveiling a Cutting-Edge Dataset for Oil Tank Detection: YOLO Models Put to the Test
Chehade, A., & Rizk, M. (2024) · International Conference on Smart Systems and Power Management (IC2SPM) · doi:10.1109/ic2spm62723.2024.10841353 · Cited by 3
“Traditional methods like the Hough Transform, Template Matching, and Histogram of Oriented Gradients (HOG) [11] rely on handcrafted features.”
PFRNet: A Small Object Detection Method Based on Parallel Feature Extraction and Attention Mechanism
Lin, H., Ji, W., & Li, J. (2025) · IEEE Access · doi:10.1109/access.2025.3538608 · Cited by 3
“Among them, object detection [7] as a key technology can accurately identify objects such as vehicles, pedestrians, and crops from these image data, providing strong support for relevant decision-making.”
Xia, S., Qin, R., Yang, L., Ma, L., & Liu, Z. (2025) · Journal of Marine Science and Engineering · doi:10.3390/jmse13020340 · Cited by 3
Süme, S., Ponomarjova, K.-M., Wendt, T. M., & Rupitsch, S. J. (2025) · Journal of sensors and sensor systems · doi:10.5194/jsss-14-37-2025 · Cited by 4
An Active Object-Detection Algorithm for Adaptive Attribute Adjustment of Remote-Sensing Images
Wang, J., Zhu, F., Wang, Q., Zhao, P., & Fang, Y. (2025) · Remote Sensing · doi:10.3390/rs17050818 · Cited by 4
A method for composite activation functions in deep learning for object detection
Liao, J., Yu, C. W., Jiang, L., Guo, L., Liang, W., Li, K., & Pathan, A. K. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-03938-7 · Cited by 6
“DL-based models for object detection are mainly divided into single-stage and two-stage models [6].”
Improved YOLOv8 Underwater Object Detection Combined with CBAM
Xu, P. Y., Luo, X., & Li, J. (2024) · doi:10.1109/isdh64927.2024.00014 · Cited by 3
“Object detection algorithms based on deep learning are mainly classified into two types: single-stage and two-stage [5].”
Yu, Z. W., Huang, Z., Hou, M., Pei, J., Yan, Y., Liu, Y.-S., & Sun, D. (2025) · Neural Networks · doi:10.1016/j.neunet.2025.107394 · Cited by 3
Robotic Hand Grasping: a review focused on multiple objects manipulation
Caicedo, Y. A., & Chiang, L. E. (2025) · Journal of Machine Design and Automation Intelligence · doi:10.1515/jmdai-2024-0003 · Cited by 3
Fu, Z., Shuo, Y., cao, P., Jing, W. P., Wang, H., & Zhang, G. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.130177 · Cited by 3
Lin, C., He, Z., Qiu, Y., & Huang, Y. (2025) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2025.3557131 · Cited by 6
“In contrast, image data from camera sensors is imbued with abundant semantic details and presents a structured and uniform format [6], [7], [8], forming a complementary relationship with the point cloud.”
Metadata
- Authors
- Zou, Z., Chen, K., Shi, Z., Guo, Y., & Ye, J.
- Affiliations
- Shanghai Artificial Intelligence Laboratory (CN)
- Carleton University (CA)
- Alibaba Group (China) (CN)
- Venue
- Proceedings of the IEEE
- Published
- 2023-03
- Volume
- 111
- Issue
- 3
- Pages
- 257-276
- ISSN
- 0018-9219 (ISSN-L)
- 1558-2256
- OpenAlex
- W4320002812
- Cited by
- 3,689
- References
- 272
- Citations by year
- 2026: 610 · 2025: 1013 · 2024: 940 · 2023: 541
- FWCI
- 280.65
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- Proceedings of the IEEE (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- National Natural Science Foundation of China (62125102)
- National Key Research and Development Program of China
- Fundamental Research Funds for the Central Universities
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Visual Attention and Saliency Detection
- Keywords
- object detection, evaluation metrics, deep learning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-07, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-07 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.