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 501–600 of 3,144
Liu, D., Parmiggiani, A., & Norton, T. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.110533 · Cited by 14
Enhanced YOLO11 for lightweight and accurate drone-based maritime search and rescue object detection
Zhao, B., Zhao, J., Song, R., Yu, L., Zhang, X., & Liu, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0321920 · Cited by 19
“Object detection methods can generally be classified into two main categories: two-stage and one-stage approaches [20].”
IRS-DETR: Lightweight real-time transformer for infrared ship detection in maritime surveillance
Zheng, Y., Ao, X., Xie, W., Lv, X., Gao, Z., Liu, X., & Fang, Z. (2025) · Ocean Engineering · doi:10.1016/j.oceaneng.2025.123319 · Cited by 19
Similar Category Enhancement Network for Discrimination on Small Object Detection
Wei, Z., Zhang, T., Sun, X., Zhuang, L., Wang, D., Marinoni, A., & Gao, L. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3628045 · Cited by 26
“O BJECT detection is a fundamental task for interpreting remote sensing imagery, encompassing two principal tasks: object localization and object classification [1], [2] [3], [4] [5].”
Automated Scrap Steel Grading via a Hierarchical Learning-Based Framework
Tu, Q., Li, D., Xie, Q., Dai, L., & Wang, J. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3206816 · Cited by 17
“Thanks to the rapid development of artificial intelligence in recent years, deep learning (DL) techniques [1], [2] earn a dramatic improvement in both accuracy and efficiency for object detection and classification tasks.”
BackgroundDroMOD: A Drone-Based Multi-Scope Object Detection System
Abdellatif, T., Sedrine, M. A., & Gacha, Y. (2023) · IEEE Access · doi:10.1109/access.2023.3253767 · Cited by 16
“This prevents the use of heavy-algorithms such as real time DL algorithms for multi-object detection [2] and complex and convolutional neural network models [3].”
MethodTian, Y. M., Zhang, J., Yin, Z., & Dong, H. (2023) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52729.2023.00861 · Cited by 31
“These center-based methods [17, 23, 47] have succeeded in simplicity and speed.”
MethodBackground Instance-Based Copy-Paste Data Augmentation for Object Detection
Zhang, L., Xing, Z., & Wang, X. (2023) · Electronics · doi:10.3390/electronics12183781 · Cited by 14
Jun, L., & Wang, X. (2023) · Frontiers in Plant Science · doi:10.3389/fpls.2023.1255119 · Cited by 17
Pieczyński, D., Ptak, B., Kraft, M., & Drapikowski, P. (2023) · Applied Sciences · doi:10.3390/app132212401 · Cited by 23
Crop Identification and Growth Stage Determination for Autonomous Navigation of Agricultural Robots
Cortinas, E., Emmi, L., & de Santos, P. G. (2023) · Agronomy · doi:10.3390/agronomy13122873 · Cited by 15
Environmental Effects on DNN Performance During the Autonomous Refueling Task
Civetta, C. A., Miller, D., Mwaffo, V., & Costello, D. H. (2024) · AIAA SciTech Forum · doi:10.2514/6.2024-2901 · Cited by 15
Huo, L., Hou, J., Feng, J., Wang, W., & Liu, J. (2024) · Remote Sensing · doi:10.3390/rs16040624 · Cited by 14
“It usually outputs a bounding box around the detected object and the corresponding class label [10].”
BackgroundAPNet: Accurate Positioning Deformable Convolution for UAV Image Object Detection
Zhang, P., Zhang, G., & Yang, K. (2024) · IEEE Latin America Transactions · doi:10.1109/tla.2024.10472961 · Cited by 16
“[10] considers DETR as a one-stage detector.”
YOLOMH: you only look once for multi-task driving perception with high efficiency
Liu, F., Sun, B., Miao, J., & Su, W. (2024) · Machine Vision and Applications · doi:10.1007/s00138-024-01525-3 · Cited by 15
“The recently proposed YOLOv8 absorbs the excellent ideas of ELAN [13] in yolov7 in the backbone network, and replaces the C3 structure with the gradient-rich C2F structure to achieve further lightweighting [4].”
Accelerated Neural Enhancement for Video Analytics With Video Quality Adaptation
Mi, L., Yuan, T., Wang, W., Dai, H., Sun, L., Zheng, J., Chen, G., & Fu, X. (2024) · IEEE/ACM Transactions on Networking · doi:10.1109/tnet.2024.3375108 · Cited by 21
“Although DNN-aware video enhancement provides a promising way to improve inference accuracy [16], there is still much room for improvement.”
BackgroundZhang, J., Chen, Z., Zou, H., Xue, S., He, J., & Li, J. (2024) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2024.2338921 · Cited by 13
“Deep learning based on convolutional neural networks (CNNs) has been widely used in object detection due to its high precision and adaptability [4].”
Zhang, W., Zhang, X., Xu, X., Xu, Y., Shao, Z., Shi, J., Wei, S., & Zeng, T. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3383870 · Cited by 15
“The most noteworthy thing is that considering the rapid development and progress in the field of object detection [41], we did not restrict the detector when designing GNN-JFL.”
MethodAutomatic grain unloading method for track-driven rice combine harvesters based on stereo vision
Cui, Z., Hu, J., Yu, Y., Cao, G., Zhang, H., Chai, X., Chen, H., & Xu, L. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.108917 · Cited by 15
CoLLaRS : A cloud–edge–terminal collaborative lifelong learning framework for AIoT
Hu, S., Lin, J., Lu, Z., Du, X., Duan, Q., & Huang, S.-C. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.04.046 · Cited by 15
Jiang, H., Fu, W., Zhang, X., & Zhang, C. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad480c · Cited by 14
“In recent years, deep learning techniques have achieved unprecedented success in numerous advanced computer vision tasks including image classification [10], image segmentation [11], 3D reconstruction [12], and object detection [13].”
Naito, H., Shimomoto, K., Fukatsu, T., Hosoi, F., & Ota, T. (2024) · AgriEngineering · doi:10.3390/agriengineering6020106 · Cited by 15
“Over the past two decades, the technology for detecting particular objects in images, including fruits, has progressed dramatically in accuracy and speed since the advent of deep learning [5].”
BackgroundSun, S., Chen, C., Yang, B., Yan, Z., Wang, Z., He, Y., Wu, S., Li, L., & Fu, J. (2024) · Drones · doi:10.3390/drones8070299 · Cited by 21
Bickel, S., Goetz, S., & Wartzack, S. J. (2024) · Applied Sciences · doi:10.3390/app14146106 · Cited by 17
Tian, Z., Liu, H., Wu, J., Chen, W., Zheng, R., & Wang, Z. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3435130 · Cited by 14
“Deep learning-based target detection is currently the main-stream method in this field, which can be divided into two-step detection and one-step detection, the former can be regarded as a “coarse-to-fine” algorithm, and the latter can be regarded as a “one-step” algorithm [18].”
CFFM: Multi-task lane object detection method based on cross-layer feature fusion
Zhang, Y., Zheng, Y., Tu, Z., Wu, C., & Zhang, T. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.125051 · Cited by 16
Hui, J., Zhang, Y., Ding, K., Guo, L., Chen, C., & Wang, L. (2024) · Advanced Engineering Informatics · doi:10.1016/j.aei.2024.102778 · Cited by 15
A Novel Detector for Wind Turbines in Wide-Ranging, Multiscene Remote Sensing Images
Xie, J., Tian, T., Hu, R., Yang, X., Xu, Y., & Zan, L. (2024) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2024.3460730 · Cited by 14
“Theimpressiveevolutionofdeep learning-basedobjectdetectionalgorithmsspansvariousstages, encompassingtwo-stage,one-stage,anchor-based,anchor-free, andkeypointdetectionmethodologies[16],[17].”
Real-Time Detection of Insulator Defects with Channel Pruning and Channel Distillation
Meng, D., Xu, X., Jiang, Z., & Xu, L. (2024) · Applied Sciences · doi:10.3390/app14198587 · Cited by 13
Bilyk, T., Goryaeva, A. M., Marinica, M., Flament, C., Sabathier, C., Leroy, É., Loyer-Prost, M., & Meslin, E. (2024) · Scientific Reports · doi:10.1038/s41598-024-74894-4 · Cited by 14
TTSDA-YOLO: A Two Training Stage Domain Adaptation Framework for Object Detection in Adverse Weather
Zhang, M., Rong, Q., & Jing, H. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3497132 · Cited by 18
“In recent years, with the rapid development of deep learning, new advancements have been made in object detection, leading to significant breakthroughs and making it a prominent research hotspot [27].”
Advances in artificial intelligence: a review for the creative industries
Anantrasirichai, N., Zhang, F., & Bull, D. (2026) · Artificial Intelligence Review · doi:10.1007/s10462-026-11494-w · Cited by 107
A Survey on Real-Time Object Detection on FPGAs
Hozhabr, S. H., & Giorgi, R. (2025) · IEEE Access · doi:10.1109/access.2025.3544515 · Cited by 25
“Although they usually suffer from poor accuracy, especially in detecting small objects, compared to two-stage detectors [30], one-stage detectors feature simpler architectures, higher speed, and adaptability to various scales [29].”
InfluentialGPRTransNet: A deep learning–based ground-penetrating radar translation network
Lu, G., Kou, L., Niu, P., Lv, G., Zhang, X., Liu, J., & Xie, Q. (2025) · Tunnelling and Underground Space Technology · doi:10.1016/j.tust.2025.106557 · Cited by 16
Application of machine learning in LC-MS-based non-targeted analysis
Jin, Z.-L., Chen, L., Wang, Y., Shi, C.-T., Zhou, Y., & Xia, B. (2025) · TrAC Trends in Analytical Chemistry · doi:10.1016/j.trac.2025.118243 · Cited by 20
A systematic review: object detection
Guleria, A., Varshney, K., Garima, G., & Jindal, S. (2025) · AI & Society · doi:10.1007/s00146-025-02372-0 · Cited by 21
An advanced deep learning method for pepper diseases and pests detection
Wang, X., Liu, J., & Chen, Q. (2025) · Plant Methods · doi:10.1186/s13007-025-01387-4 · Cited by 19
Liu, D., Zhang, J., Li, T., Qi, Y., Wu, Y., & Zhang, Y. (2023) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2023.3292890 · Cited by 13
“traditional methods and deep learning-based methods [1].”
Yao, G., Zhou, W., Liu, M., Xu, Q., Wang, H., Li, J., & Ju, Y. (2021) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2021.3132416 · Cited by 17
Traffic-Net: 3d Traffic Monitoring Using a Single CameraPreprint
Rezaei, M., Azarmi, M., & Mir, F. M. P. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4152596 · Cited by 18
Deo, R., John, C. M., Zhang, C., Whitton, K., Salles, T., Webster, J. M., & Chandra, R. (2024) · Scientific Data · doi:10.1038/s41597-024-03766-3 · Cited by 15
“Deep learning is a machine learning methodology that has been prominently used in computer vision problems such as image (object) classification 25 , and also been used for classifying underwater marine environments 26–28 .”
MethodOriented Object Detection in Optical Remote Sensing Images using Deep Learning: A SurveyPreprint
Wang, K., Wang, Z., Li, Z., et al. (2024) · Research Square · doi:10.21203/rs.3.rs-5346546/v1 · Cited by 13
Raza, S. M., Abidi, S. M. H., Masuduzzaman, M., & Shin, S. Y. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131357 · Cited by 13
Iterative knowledge distillation and pruning for model compression in unsupervised domain adaptation
Wang, Z., Shi, L. X., Mei, Z., Zhao, X. H., Wang, Z., & Li, J. (2025) · Pattern Recognition · doi:10.1016/j.patcog.2025.111512 · Cited by 12
Zeinaty, C. E., Hamidouche, W., Herrou, G., & Ménard, D. (2025) · ACM Computing Surveys · doi:10.1145/3744339 · Cited by 36
“A detailed examination of novel optimization strategies, tailored to the unique constraints of MCUs with limited power and memory, is provided, bridging the gap between theoretical understanding and practical deployment in edge AI applications [35].”
YOLO-UIR: A Lightweight and Accurate Infrared Object Detection Network Using UAV Platforms
Wang, C., Wang, R., Wu, Z., Bian, Z., & Huang, T. (2025) · Drones · doi:10.3390/drones9070479 · Cited by 18
YOLO-Extreme: Obstacle Detection for Visually Impaired Navigation Under Foggy Weather
Wang, W., Jing, B., Yu, X., Zhang, W., Wang, S., Tang, Z., & Yang, L. (2025) · Sensors · doi:10.3390/s25144338 · Cited by 14
ESOD-YOLOv8: Small object detection enhanced with auto-disturbance rejection convolution
Yu, Z., Liang, H., Ye, O., & Zhang, Y. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129046 · Cited by 27
Zhou, G., Xu, Q., Liu, Y., Liu, Q., Ren, A., Zhou, X., Li, H., & Shen, J. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3602640 · Cited by 21
“However, both approaches typically demand extensive computational resources and high-performance hardware, limiting their practical utility and applicability [20].”
Social robot navigation: a review and benchmarking of learning-based methods
Alyassi, R., Cadena, C., Riener, R., & GRANADOS, D. F. P. (2025) · Frontiers in Robotics and AI · doi:10.3389/frobt.2025.1658643 · Cited by 25
Improved YOLOv9-based remote sensing image detection method
Chen, M., Wang, C., Yu, Y. K., Li, Y., Li, Y., & Chen, X. (2025) · Scientific Reports · doi:10.1038/s41598-025-28528-y · Cited by 12
High-Precision Detection Algorithm for Metal Workpiece Defects Based on Deep Learning
Xu, X., Zhang, G., Zheng, W., Zhao, A., Zhong, Y., & Wang, H. (2023) · Machines · doi:10.3390/machines11080834 · Cited by 14
Automatically Prepare Training Data for YOLO Using Robotic In-Hand Observation and Synthesis
Chen, H., Wan, W., Matsushita, M., Kotaka, T., & Harada, K. (2023) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2023.3304420 · Cited by 18
“The training process was stopped when the mAPs (mean Average Precision) [47] for all objects reached higher than 99 .”
BackgroundCrowdDCNN: Deep convolution neural network for real-time crowd counting on IoT edge
Chavan, R., kanamarlapudi, A., Rani, G., Thakkar, P., & Dhaka, V. S. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107089 · Cited by 14
An ISAR Image Component Recognition Method Based on Semantic Segmentation and Mask Matching
Zhu, X., Zhang, Y., Wang, L., Fang, Y., & He, J. (2023) · Sensors · doi:10.3390/s23187955 · Cited by 13
“In semantic segmentation research, using local information is beneficial to improve the classification accuracy at the pixel level, while introducing global information is a key way to solve the local ambiguity problem.”
Fu, M., Wang, Z., Wang, J., Wang, Q., Ma, Z., & Wang, D. (2023) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2023.3313641 · Cited by 13
Wang, H., Salunkhe, O., Quadrini, W., Lämkull, D., Ore, F., Johansson, B., & Stahre, J. (2023) · Procedia CIRP · doi:10.1016/j.procir.2023.09.127 · Cited by 14
Improved YOLOv5-based image detection of cotton impurities
Hu, D., Liu, X., & Xu, J. (2024) · Textile Research Journal · doi:10.1177/00405175231221296 · Cited by 12
“8 YOLOv5 input is augmented with Mosaic data from YOLOv4 to improve the training speed and accuracy of the model.”
MethodZero-guidance Segmentation Using Zero Segment Labels
Rewatbowornwong, P., Chatthee, N., Chuangsuwanich, E., & Suwajanakorn, S. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.00113 · Cited by 27
Li, J., Liu, Z., & Wang, D. (2024) · Agriculture · doi:10.3390/agriculture14020273 · Cited by 14
“Many experiments have shown that the one-stage detection model has slightly lower detection accuracy, but its detection speed is faster than that of the two-stage detection model [18–23].”
ResultReview of current vision-based robotic machine-tending applications
Jia, F., Ma, Y., & Ahmad, R. (2024) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-024-13168-9 · Cited by 17
“…vision domain, such as COCO, MNIST, and ICDAR, it is difficult to directly apply these datasets to specific manufacturing cases because these benchmarks are mostly created for particular tasks such as vehicles detection, pedestrians detection, car plate recognition, and traffic sign detection [76].”
InfluentialGuang, J., Hu, Z., Wu, S., Zhang, Q., & Liu, J. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.123497 · Cited by 16
Dynam-LVIO: A Dynamic-Object-Aware LiDAR Visual Inertial Odometry in Dynamic Urban Environments
Shi, J., Wang, W., Qi, M., Li, X., & Ye, Y. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3395320 · Cited by 16
“However, most lightweight object detection networks use one-stage detector [28], resulting in the inevitable occurrence of false and missed detections.”
BackgroundOATracker: Object-aware anti-occlusion 3D multiobject tracking for autonomous driving
Zhang, X., Tan, X., An, Y., Li, Y., & Fan, Z. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124158 · Cited by 15
Zhu, R., Ma, S., Lian, J., He, L., & Mei, S. (2024) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2024.3399780 · Cited by 14
“This has greatly contributed to significant advancements in remote sensing studies [1], [2], [3], [4], such as object detection [5], [6], [7], scene classification [8], [9], [10], and object tracking [5], [11], [12], [13].”
Quantitative Analysis of Deep Learning-Based Object Detection Models
Elgazzar, K., Mostafi, S., Dennis, R., & Osman, Y. (2024) · IEEE Access · doi:10.1109/access.2024.3401610 · Cited by 13
“As such, in the last decade, a large number of surveys have been published focusing primarily on object detection based on deep learning [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30].”
BackgroundYu, L., Zhi, X., Zhang, S., Jiang, S., Hu, J., Zhang, W., & Huang, Y. (2024) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2024.3403548 · Cited by 17
GraspUI: Seamlessly Integrating Object-Centric Gestures within the Seven Phases of Grasping
Sharma, A., Ivanov, A., Lai, F., Grossman, T., & Santosa, S. (2024) · Designing Interactive Systems Conference · doi:10.1145/3643834.3661551 · Cited by 17
“Our interactive prototype lays the technical groundwork for the GraspUI gesture recognition architecture, overcoming challenges with simultaneous un-instrumented object and hand tracking [29, 82].”
BackgroundAdvancements and Challenges in Hand Gesture Recognition: A Comprehensive Review
Murad, B. K., & Alasadi, A. H. H. (2024) · Iraqi Journal for Electrical And Electronic Engineering · doi:10.37917/ijeee.20.2.13 · Cited by 15
“Local topological descriptors match the model to the image’s edges, while a shape context descriptor, introduced by Zou et al [39], distinguishes the position of a specific point on a shape.”
CTOD: Cross-Attentive Task-Alignment for One-Stage Object Detection
Yao, R., Rong, Y., Huang, Q., & Xiong, S. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3422879 · Cited by 13
Multi-source-free Domain Adaptive Object Detection
Zhao, S., Yao, H., Lin, C., Gao, Y., & Ding, G. (2024) · International Journal of Computer Vision · doi:10.1007/s11263-024-02170-z · Cited by 21
Collaborative Inference in DNN-Based Satellite Systems with Dynamic Task Streams
Guan, J., Zhang, Q., Murturi, I., Donta, P. K., Dustdar, S., & Wang, S. (2024) · International Conference on Communications · doi:10.1109/icc51166.2024.10622590 · Cited by 22
“Many machine learning algorithms, especially Deep Neural Networks (DNN), are employed for the inference of raw image observation applications such as AI-enabled forest fires [3].”
BackgroundHe, Y. G., Cai, C., Chen, G., Hu, J., Hu, S., & Fu, J. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.125741 · Cited by 14
Li, M.-D., Xiao, S., & Chen, S.-W. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3533930 · Cited by 12
“Four commonly used evaluation criteria [43] are employed to assess the accuracy and completeness of the component recognition tasks, including recall, precision, F1 score, and mean average precision (mAP). indicates the missed target rate, while precision can represent the false alarm ratio.”
Peng, R., Liao, C., Pan, W., Gou, X., Zhang, J., & Lin, Y. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.129844 · Cited by 17
Mask R-CNN for Geospatial Object Detection
AL-Alimi, D., Shao, Y., Alalimi, A., & Abdu, A. (2020) · International Journal of Information Technology and Computer Science · doi:10.5815/ijitcs.2020.05.05 · Cited by 16
Du, L., Chen, X., Pei, Z., et al. (2022) · Journal of Advanced Transportation · doi:10.1155/2022/2259113 · Cited by 12
“Many deep learning-based object detection methods were proposed in recent years, including one-stage and two-stage detection methods, as shown in Figure 1 [7].”
Li, C., & Lang, H. (2022) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2022.3206247 · Cited by 14
“With the rapid development of deep learning, object detection based on convolutional neural network (CNN) has shined in the field of natural images [21], [22].”
BackgroundAutomotive Object Detection via Learning Sparse Events by Spiking Neurons
Zhang, H., Li, Y., Leng, L., Che, K., Liu, Q., Guo, Q., Liao, J., & Cheng, R. (2024) · IEEE Transactions on Cognitive and Developmental Systems · doi:10.1109/tcds.2024.3410371 · Cited by 24
“Addressing this gap, the recently advanced dynamical vision sensor (DVS), also known as. the event-based sensor [3], [4], [5], [6], [7], [8], offers a compelling alternative.”
Wei, Y., Chen, Q., & Wei, X. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3573015 · Cited by 11
“Space target detection refers to the process of identifying, tracking, and locating targets in space by using cameras or radars [4], [5].”
A Frequency Domain-Enhanced Transformer for Nighttime Object Detection
Li, Y., & Shen, L. (2025) · Sensors · doi:10.3390/s25123673 · Cited by 16
An Overview on Computer Vision Analysis in the Airport Applications
Talbi, A., Abdellaoui, M., Pons‐Prats, J., & Kuljanin, J. G. (2025) · IEEE Access · doi:10.1109/access.2025.3580299 · Cited by 19
Image Quality Assessment: From Human to Machine Preference
Li, C., Tian, Y., Ling, X., Zhang, Z., Duan, H., Wu, H., Jia, Z., Liu, X., Min, X., Lu, G. S., et al. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00709 · Cited by 45
“Thus, we consider the detection failed when the categories are different, and only further calculate the IoU when they are consistent [96], as described in S DET : where C dis,ref denotes the category sequence ordered by confidence, and B dis,ref is the detected bounding box.”
Remote sensing and image analysis of macro-plastic litter: A review
Kako, S., Kataoka, T., Matsuoka, D., Takahashi, Y., Hidaka, M., Aliani, S., Andriolo, U., Dierssen, H. M., van Emmerik, T. H. M., Gonçalves, G. R., et al. (2025) · Marine Pollution Bulletin · doi:10.1016/j.marpolbul.2025.118630 · Cited by 20
Adam, M., Song, J., Yu, W., & Li, Q. (2025) · Future Internet · doi:10.3390/fi17090431 · Cited by 15
“Object detection algorithms are broadly classified into two categories based on the number of times the network processes the input image: two-stage and one-stage detectors [15].”
What Is in the Frequency: Wavelet-Guided Semantic Understanding for Infrared Small Target Detection
Guo, W. Q., Cai, F., & Quan, W. (2025) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2025.3623373 · Cited by 11
“Infrared small targets often appear as point-like or occupy only a few pixels, typically exhibiting low signal-to-noise ratios and weak texture information [6], [7].”
A Review of Key Technologies and Recent Advances in Intelligent Fruit-Picking Robots
Lin, T., Sun, F., Li, X., Guo, X., Ying, J., Wu, H., & Li, H. (2026) · Horticulturae · doi:10.3390/horticulturae12020158 · Cited by 16
Automatic terminal guidance for small fixed‐wing unmanned aerial vehicles
Yang, Y., Yang, Y., Li, J., et al. (2022) · Journal of Field Robotics · doi:10.1002/rob.22113 · Cited by 11
“Its main idea is to predefine a set of reference boxes (also known as anchor boxes) with different sizes and aspect ratios at different locations of an image, and then predict the detection box based on these references (Zou et al., 2019).”
BackgroundDFEN: Dual Feature Enhancement Network for Remote Sensing Image Caption
Zhao, W., Yang, W., Chen, D., & Wei, F. (2023) · Electronics · doi:10.3390/electronics12071547 · Cited by 16
“Because of advances in remote sensing technology, computer vision tasks such as scene classification and object detection [2] can now obtain scene categories and ground objects from remote sensing images, but not semantic relationships among ground objects.”
BackgroundAReNet: Cascade learning of multibranch convolutional neural networks for human activity recognition
Boudjema, A., Titouna, F., & Titouna, C. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-17496-y · Cited by 14
“The first part extracts useful features using a bunch of convolutional and pooling operations [42] while the second part, the multilayer per-ceptron, contains a set of fully connected layers where every neuron in one layer is coupled to every neuron in the next layer to perform classification.”
Li, G., Zhao, S., Li, M., Zhou, M., & Ying, Z. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107677 · Cited by 14
Yuan, W. (2024) · Data · doi:10.3390/data9020036 · Cited by 37
Abdulkareem, I. M., AL-Shammri, F. K., Khalid, N. A. A., & Omran, N. A. (2024) · International Journal of Online and Biomedical Engineering (iJOE) · doi:10.3991/ijoe.v20i05.47171 · Cited by 13
TIDE: Test-Time Few-Shot Object Detection
Li, W., Wei, H., Wu, Y., Yang, J., Ruan, Y., Li, Y., & Tang, Y. G. (2024) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2024.3371699 · Cited by 13
“Over the past decade, substantial progress has been made in object detection utilizing deep learning techniques [6], [7], [8], [9], [10], [11].”
BackgroundAn, R., Liu, G., Qian, Y., Xing, M., & Tang, H. (2024) · Infrared Physics & Technology · doi:10.1016/j.infrared.2024.105343 · Cited by 11
LMFE-RDD: a road damage detector with a lightweight multi-feature extraction network
He, Q., Li, Z., & Yang, W. (2024) · Multimedia Systems · doi:10.1007/s00530-024-01367-z · Cited by 13
“Object detection is an important computer vision task that detects instances of visual objects of a certain class, such as humans, animals, or cars, in digital images [1].”
Infrastructure-assisted 3D detection networks based on camera-lidar early fusion strategy
Yao, J., Zhou, J. G., Wang, Y., Gao, Z., & Hu, W. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128180 · Cited by 14
Single-stage zero-shot object detection network based on CLIP and pseudo-labeling
Li, J., Sun, S., Zhang, K., Zhang, J., & Zhuo, L. (2024) · International Journal of Machine Learning and Cybernetics · doi:10.1007/s13042-024-02321-1 · Cited by 11
The hawk eye scan: Halyomorpha halys detection relying on aerial tele photos and neural networks
Palazzetti, L., Rangarajan, A. K., Dinca, A., Boom, B., Popescu, D., Offermans, P., & Pinotti, C. M. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109365 · Cited by 11
Wang, G., Zhao, H., Lyu, S., Cheng, G., Chang, Q., Feng, W., Zhao, Q., & Shi, Z. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3452010 · Cited by 19
Self-citation
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.