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 3,001–3,100 of 3,144
Anomaly Detection and Recognition in Complex Commercial Kitchen Environments
Yang, S., Wang, Q., Zhou, Y., & Lai, H. (2026) · IEEE Access · doi:10.1109/access.2026.3717751
“One-stage detectors, such as SSD [12] and the YOLO series [13], [14], directly predict object categories and bounding boxes in an end-to-end manner.”
Real-time Fish Detection and Counting with YOLOv11
Nghia, L. V., Vu, Q., & Ronzhin, A. L. (2026) · Smart innovation, systems and technologies · doi:10.1007/978-3-032-17407-9_2
Neuromorphic Object Detection: An In-Depth Study and Future Directions
Li, J., Li, D., Glover, A., Xiaopeng Fan, B., Li, G., Bartolozzi, C., Benosman, R. B., & Tian, Y. (2026) · Proceedings of the IEEE · doi:10.1109/jproc.2026.3717677 · Cited by 1
“Object detection [1], [2], [3] is a fundamental and long-standing topic with a wide range of applications, such as agile robotics [4], [5], autonomous driving [6], surveillance [7], and augmented reality [8].”
Pang, J., Shen, Q., Ye, Z., Cai, Z., & Li, Y. (2026) · Reliability Engineering & System Safety · doi:10.1016/j.ress.2026.113224
Huang, Z., Chong, J., Li, M., Qi, X., & Yang, Z. (2026) · IEEE Internet of Things Journal · doi:10.1109/jiot.2026.3718161
“UAVs integrated with visual cameras [5], LiDAR [6], or thermal imaging systems [7], perform e ff ectively under normal conditions.”
Image-Based Visual Detection System for Assembly Defects of Electric Operating Handles
朱, 海. (2026) · Journal of Image and Signal Processing · doi:10.12677/jisp.2026.153035
Lightweight Unsupervised Domain Adaptation Framework for Mass Detection in Mammography
Nguyen, K.-T., Pham, B.-T., Nguyen, C.-N., Shih, H.-C., Le, T.-P., & Wang, J. (2026) · IEEE Access · doi:10.1109/access.2026.3718017
SODBoost: Density-Guided Zoom-In Mosaic for Small Object Detection
余金仑, Chen, Y., Shi, Y., & Yang, J. (2026) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv66570.2026.11624034
“Despite remarkable progress in generic object detection [2], small objects remain notably challenging due to weak feature representation and unstable optimization: tiny instances are easily submerged within complex backgrounds and provide limited cues on high-stride feature maps, suppressing…”
Automated Litter Characterization for Smart Cities
Diaz-Silva, E., Gao, G., Begur, N., Charles, S., Arra, R., Chatterjee, S., Zarif, M., & Hallstrom, J. O. (2026) · IEEE International Conference on Smart Computing Workshops and Other Affiliated Events (SMARTCOMP Companion) · doi:10.1109/smartcomp-companion70724.2026.00059
“Zou et al. [9] provide a comprehensive survey highlighting the impact of convolutional neural networks (CNNs).”
Established Deep Learning Object Detectors Remain Adequate for Road Infrastructure Monitoring
Doulamis, A. D., Katsamenis, I., Protopapadakis, E. E., et al. (2026) · Technical Annals · doi:10.12681/ta.45596
“Object detection is especially practical among DL tasks: it simultaneously localizes defects with bounding boxes and assigns class labels, striking a useful balance between spatial detail and computational efficiency [10].”
Edge-Based PPE Detection for Urban Construction Under Realistic Visual Conditions
Miglionico, G. C., Rienzo, F. D., Ducange, P., Marcelloni, F., & Vallati, C. (2026) · IEEE International Conference on Smart Computing Workshops and Other Affiliated Events (SMARTCOMP Companion) · doi:10.1109/smartcomp-companion70724.2026.00064
“YOLOv8 adopts an anchor-free formulation, directly regressing bounding box parameters from feature map locations, and employs decoupled branches for classification and localization [5], [18].”
FaceMixup: Enhancing Facial Expression Recognition through Mixed Face Regularization
Souza, M., Faria, F. A., Texeira, R., & Segundo, M. P. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31933-3_46 · Cited by 1
Hyper Look-Ahead Network for Remote Sensing Object Detection
Jiang, L., Li, Y., & Bai, T. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3716854
“Currently, object detection algorithms for remote sensing images based on deep learning can be categorized into three main types: supervised, weakly supervised, and unsupervised methods [2], [3], [4].”
YOLO-Driven Vessel Detection and Multi-Object Tracking in Fixed-Camera Marina Environments
Lopac, N., Severinski, K., Grubišić, N., et al. (2026) · Technologies · doi:10.3390/technologies14080482
Comparison of Real-Time Multi-object Tracking with Limited Hardware Resources
Bernhart, C., Strohmayer, J., Kampel, M., Peer, M., & Kleber, F. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-31654-7_38
Kaveti, K. K., Akhila, J. V., & Divya, N. (2026) · doi:10.1109/icdca69396.2026.11620423
“In practical clinical environments, full microscopic images must be analyzed directly, which limits the applicability of classification-only approaches [9], [25], [28].”
Wang, P., Wang, J., Shao, J., et al. (2026) · International Conference on Control and Automation · doi:10.1109/icca69928.2026.11618038
“While numerous state-of-the-art object detection models exist [12], [13], SAR detection task is primarily focused on locating single-class targets (i.e., humans) in a resource-constrained environment such as drones, which calls for a lightweight model that can run in real-time without sacrificing…”
NECTAR: novel ensemble comb targeting & automated reporting
Yılmaz, E., & ALTINEL, A. B. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116801
A Real-Time Vehicle Detection and Re-ID System from the Drone Perspective
Wang, Y., Liu, Y., Wang, Z., et al. (2026) · Communications in computer and information science · doi:10.1007/978-981-95-9938-7_4
Polarization representation spaces for enhancing target detection and recognition: A review
Zang, S., Ding, H., Gao, X., & Guo, Z. (2026) · Defence Technology · doi:10.1016/j.dt.2026.07.028
Chang, Y. L., Zhao, X., Li, Y., et al. (2026) · Concurrency and Computation Practice and Experience · doi:10.1002/cpe.70893
ALSDet: a global context-enhanced network for detecting small-target diseases on apple leaves
Gong, X., Xu, P., Ren, R., Gong, F., & Zhang, S. (2026) · Frontiers in Plant Science · doi:10.3389/fpls.2026.1883033
“Object detection algorithms have been signi fi cantly improved by recent advances in deep learning (Sharma and Mir, 2020; Zou et al., 2023; Wu et al., 2020).”
Bao, H., Huang, Y., Wang, Y., Ye, J., Wang, X., Chen, X., Zhao, Y., Zhou, T., Elhoseiny, M., & Zhang, X. (2026) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3770855.3817506 · Cited by 1
“Large Vision–Language Models (LVLMs) have rapidly advanced multimodal perception and reasoning [21], enabling unified image–text understanding across tasks such as VQA, grounding, and GUI manipulation [1, 3, 8, 52].”
Self-citationEnd-to-End NMS-Free Detection for High-Density Non-Rigid E-Bikes via Parameter-Free 3D Attention
Yang, Z., Huang, Q., Gao, S., et al. (2026) · doi:10.1109/dlcv69906.2026.11635726
“However, accurately detecting such objects under severe physical occlusion remains a formidable computer vision challenge [2].”
Hung, W.-C., Wu, C.-S., Chang, C.-M., et al. (2026) · Biomedicines · doi:10.3390/biomedicines14081786
“To address this limitation, we explored object detection models for automated tumor region cropping [8,9].”
Consistencydet: few-step self-consistent box denoising for efficient object detection
Jiang, L., Wang, Z., Wang, C., Li, M., & Leng, J. (2026) · The Visual Computer · doi:10.1007/s00371-026-04674-w
“Object detection predicts object locations and categories in images [1–3], supporting tasks such as instance segmentation, pose estimation, and action recognition [4–6].”
-, Z. W., Fang, J., Fang, T., et al. (2026) · KSCE Journal of Civil Engineering · doi:10.1016/j.kscej.2026.100705
Detector-Guided Multi-View Visual Monitoring of Bolt Loosening in Hydropower Generator Rotors
Lyu, J., Guo, J., Yuan, F., Ran, Y., Gong, H., Wu, T., & Zhang, T. (2026) · Applied Sciences · doi:10.3390/app16167930
Hyperspectral Object Detection via Spectral Prototype Mining and Spatial-Spectral Mamba Fusion
Wang, J., Li, Z., Wang, Y., Fu, G., Xing, C., & Xue, J. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3722059
“Most existing detection frameworks primarily rely on color imagery and identify objects using spatial visual cues, such as structural patterns, texture characteristics, edge information, and color contrasts [6].”
VID-YOLO: A Multi-Scale Risk-Aware Object Detection Method for Visually Impaired Navigation
Li, X., Yang, Q., Li, J., et al. (2026) · International Conference on Electronic Information Technology and Computer Engineering (EITCE) · doi:10.1109/eitce70137.2026.11634625
“Object detection has developed from two-stage detectors to single-stage detectors that balance accuracy and speed [2-4] .”
YOLO-SWM: an improved YOLO11 detection method for multiple vehicle targets
Zhai, H., Zhao, X., Liu, H., & Shi, P. (2026) · IEEE China Conference on System Simulation Technology and its Applications (CCSSTA) · doi:10.1109/ccssta69471.2026.11635115
“Object detection represents one of the most critical and difficult problems in computer vision, which aims to locate the position and identify the category of the objects in the given image [1].”
Zhang, T., Liu, X., Li, J., Yu, H., & Liu, Z. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3722209
“However, existing deep learning methods primarily focus on object detection and recognition, while they can identify objects in images and answer the question “What objects are present in the image,” their capabilities are limited to basic object perception [7], [8].”
LASOD-YOLO: A Lightweight Global Context Modeling for Aerial Small-Object Detection
Wang, Z., & Huang, J. X. (2026) · Proceedings - International Conference on Image Processing · doi:10.1109/icip61757.2026.11630018
TIBER-YOLO: an improved lightweight model for underwater object detection
ANTWI, S. A., Amoako, J. Y., & Enyan, M. (2026) · Intelligent Marine Technology and Systems · doi:10.1007/s44295-026-00111-9
“The advent of DLOD has substantially improved the accuracy and efficiency of object detection methods (Zou et al. 2023).”
A Review of Robotics for Unmanned Machining
Khishtan, A., Khoshnazar, M., Rong, Z., & Lee, J. (2026) · International Journal of Precision Engineering and Manufacturing · doi:10.1007/s12541-026-01607-4
A lightweight weld quality inspection method for ship structures via UAV-robot collaborative systems
Xie, Y., Cui, J., Zhang, D., Bie, A., & Song, F. (2026) · Ocean Engineering · doi:10.1016/j.oceaneng.2026.127546
Heterogeneous edge AI orchestration: A systematic review, taxonomy and future directions
Ali, B., Gill, S. S., Cuadrado, F., & Uhlig, S. (2026) · Computer Science Review · doi:10.1016/j.cosrev.2026.101044 · Cited by 1
Visual detection of flap tears on abrasive flap wheels using deep learning
Kähler, F., Ehrbar, J. G., & Schüppstuhl, T. (2026) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-026-18916-7
Ma, F., Gao, J., Pang, H., Yan, D., Bai, B., Pei, Y., & Li, S. (2026) · Journal of Computing in Civil Engineering · doi:10.1061/jccee5.cpeng-7853
DFE-RFC: Degradation-Robust Feature Network for Object Detection in Adverse Weather
Sun, H., Li, J., Huang, Y., Han, J., Tang, Y., Cai, D., & Zhang, Z. (2026) · IEEE International Conference on Mechatronics and Automation (ICMA) · doi:10.1109/icma69663.2026.11647346
“It is a fundamental task in the field of computer vision and has been widely applied in scenarios such as autonomous driving, intelligent surveillance, and industrial vision inspection [1], [2].”
Zhang, M., Du, D., Jing, H., et al. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3724296
“I NFRARED small target detection (IRSTD) is important for remote-sensing applications, including maritime surveillance, early warning systems, and environmental monitoring [1], [2], [3].”
Robust Real-Time Pig Detection for Commercial Swine Barns Using an Improved MSCA-RTDETR Model
Hao, W., Chen, Y., Xu, S., et al. (2026) · Agriculture · doi:10.3390/agriculture16161756
Deng, H., Ma, X., Guan, H., Zhu, T., Zhang, T., Zhang, Y., Gu, Z., Zhou, H., & Lu, Y. (2026) · Industrial Crops and Products · doi:10.1016/j.indcrop.2026.124141
UW-D-FINE: degradation-aware real-time underwater object detection
Ma, H., Wan, T., Dong, H., Lou, G., Feng, J., & Zhang, H. (2026) · Journal of Real-Time Image Processing · doi:10.1007/s11554-026-01964-3
“Object detection has evolved from two-stage to single-stage to Transformer-based paradigms [12].”
Huang, J., Hashim, H. B., Norman, H., et al. (2026) · PeerJ Computer Science · doi:10.7717/peerj-cs.3745
An efficient and robust object detector for complex autonomous driving scenarios
Chen, B., Xu, C., & Gu, X. (2026) · Multimedia Systems · doi:10.1007/s00530-026-02601-6
“Meanwhile, deep learn - ing algorithms have demonstrated a powerful capability superior to traditional methods in processing complex road scene images, enabling vehicles to perform detection tasks with higher efficiency [ 2].”
Shen, C., Gu, S., Mu, J., Wang, G., & Wang, M. (2026) · IET Intelligent Transport Systems · doi:10.1049/itr2.70309
AI-Based Risk Assessment System for Industrial Machine Entrapment Accidents
Jo, Y.-C., Jung, S.-K., Cha, J.-R., et al. (2026) · Journal of the Korean Society of Manufacturing Process Engineers · doi:10.14775/ksmpe.2026.25.8.115
MAP-DETR: A Frequency-Spatial State Space-Based Heavy Soil Potato Detection Algorithm
Zhang, H., Tao, J. I. N., Zhiguo, P. A. N., Yang, R., Zihao, Z. H. A. O., & Feng, L. (2026) · IEEE Access · doi:10.1109/access.2026.3725583
“In recent years, the application of computer vision in agricultural object detection has experienced a paradigm shift from manual feature engineering to deep learning [8].”
Optimization of Face-Gait Person Identification Pipelines for Edge-Deployed Cognitive Agents
Boscolo, F., Mola, G. D., & Lamberti, F. (2026) · IEEE Annual International Computer Software and Applications Conference (COMPSAC) · doi:10.1109/compsac69091.2026.00175
“One-stage detectors are widely adopted due to their favorable latency–accuracy trade-off, and recent YOLO implementations are commonly used in practical deployments [4], [5].”
Context-gated dynamic perception for small-object detection in dense aerial scenes
Gao, G., & Xie, R. (2026) · Pattern Analysis and Applications · doi:10.1007/s10044-026-01755-3
“Object detection is a fundamental visual computing task and supports applications such as autonomous driving, UAV surveillance, traffic monitoring, and remote-sensing image interpretation [1–5].”
A Multi-Camera Temporal Fusion for False Alarm Suppression in Edge-Based Weapon Surveillance
Silva, C. J. F., Del-Valle-Soto, C., Mostafa, S. M., Karim, F. K., & Aldás, J. V. (2026) · IEEE Access · doi:10.1109/access.2026.3725505
“…in both images and video streams, supporting a wide range of real-world applications The associate editor coordinating the review of this manuscript and approving it for publication was Binit Lukose . such as intelligent surveillance, smart cities, and public safety monitoring [1], [2], [3].”
Traditional ML-based analysis of anomalous crowd behavior: A thorough review
Chakole, P. D., & Satpute, V. R. (2026) · Sadhana · doi:10.1007/s12046-026-03258-7
“Object detection is the process of identifying particular visual objects within digital image data [28].”
Simpore, F., & Olayemi, M. (2026) · ASEE Annual Conference & Exposition · doi:10.18260/1-2--61004
Huang, K., Li, Y., Liu, K., et al. (2026) · IEEE International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA) · doi:10.1109/iciba68339.2026.11651683
“Of these, the AP value relates to the average precision metric, the confusion matrix in machine learning tasks and the Intersection Over Union (IOU) metric [1].”
Li, Q., Luo, H., Peng, Z., Cui, D., He, J., & Deng, X. (2026) · Expert Systems · doi:10.1111/exsy.70366
Zhang, P., Zhang, J., Liu, J., Li, X., Liu, Y., & Tan, L. (2026) · Remote Sensing · doi:10.3390/rs18172872
“Classifying these objects often relies heavily on abrupt edge transitions, corner responses, elongated contours, and subtle yet relatively stable local differences relative to the surrounding background [5,6].”
RGN-UOD: An information redistribution-guided network for underwater object detection
Wang, F., Jiang, Z., & Zhou, T. (2026) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-026-01164-w
A multi-weather-adapted object detection network for insulator defect detection
Liu, S., & Zhang, L. (2026) · Electric Power Systems Research · doi:10.1016/j.epsr.2026.114004
Detecção de objetos em cenários de escassez de rótulos usando pseudo-rótulos gerados pelo SAM3
Sobrinho, J. V. D. S., & Campista, M. E. M. (2026) · doi:10.5753/sbcup.2026.22421
SDFA: Spectral–Spatial Degradation-Aware Fusion Attention for robust underwater object detection
Dai, J., Xie, X., & Sun, Y. (2026) · Pattern Recognition · doi:10.1016/j.patcog.2026.114751
Preamble-Free Event Detection and Coarse Time-of-Arrival Estimation in LoRa LPWANsPreprint
Škrabánek, P., Pourafzal, A., & Cheffena, M. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7576576
Chiddarwar, G., Salunkhe, M. J., Joshi, Y. G., Deshmukh, R. G., Gadgune, S. Y., Nalawade, S. R., & Bongale, A. (2026) · Research Square · doi:10.21203/rs.3.rs-10381781/v1
Zhang, T., Ran, Y., Gong, H., Wu, T., Ma, T., Guo, J., & Yuan, F. (2026) · Sensors · doi:10.3390/s26175411
“These developments also reflect the broader challenge of representing small objects in modern detectors [9].”
RCB-YOLOv5s: Detection of Highway Guardrail Installation Poles Based on Improved YOLOv5s Algorithm
田, 哲. (2026) · Computer Science and Application · doi:10.12677/csa.2026.168281
Xie, Y., Lian, D., Ling, Q., Yuan, S., Wei, X., Chen, M., Zhu, D., & Cheng, X. (2026) · Journal of Hydroinformatics · doi:10.2166/hydro.2026.072
Efficient small foreign object debris detection on airport runways using RetinaNet-BiFormer-CARAFE
Chen, G., Deng, H., Di, Q., & Qiang, L. (2026) · The Journal of Supercomputing · doi:10.1007/s11227-026-08833-7
“Firstly, we select RetinaNet network as the base-line module, because it can achieve a balance between speed and accuracy based on single-stage algorithm and two-stage algorithm [9].”
InfluentialDemanov, V., Kvyatkovskaya, I. Y., & Idrissov, S. (2026) · Smart innovation, systems and technologies · doi:10.1007/978-3-032-35010-7_38
Zhang, Z., Gao, F., He, X., Lei, Y., Sun, J., Wang, J., & Hussain, A. (2026) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2026.3728463
“Among CNN-based detectors, they can be further divided into two main types: single-stage detectors and two-stagedetectors[25].”
Chen, Y., Yang, C., Yuan, C., Lu, Y., Fu, D., Fang, Y., Liu, S., & Ai, H. (2026) · Agriculture · doi:10.3390/agriculture16171884
“Advances in feature representation, multiscale modeling, and real-time inference have established a strong algorithmic basis for target recognition in complex agricultural scenes [5].”
Chiu, Y., Tsai, C., & Nisar, H. (2026) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-026-01148-w
“Affordance segmentation builds upon several foundational vision tasks, including object detection (Zou et al. 2023), semantic segmentation (Guo et al. 2018), and instance segmentation (Hafiz and Bhat 2020).”
Tracking Insects in Controlled Experiments
Zaidman, G., Aharon, S. B., Cohen, O., et al. (2026) · Agriculture · doi:10.3390/agriculture16171887
Intelligent Video Analytics for Real-Time Worker Intrusion Detection in Restricted Industrial Zones
Faiz, K., Muchtar, K., & Muchallil, S. (2026) · IEEE International Conference on Instrumentation, Networks, Future Sustainability, Information Theory, Transportation, and Systems (INFINITYS) · doi:10.1109/infinitys69204.2026.11665951
“The transition to deep learning-based object detection mitigated these limitations, introducing robust two-stage detectors such as Faster R-CNN and single-stage alternatives like SSD and the YOLO series [9].”
Kang, J.-H., Lee, S.-M., & Park, S.-H. (2026) · International Journal of Precision Engineering and Manufacturing-Green Technology · doi:10.1007/s40684-026-00926-1
Shi, H., Wang, Y., Yao, J., & Yu, W. (2026) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2026.3729008
“Deep learning-based object detection algorithms can be divided into two-stage [31] and one-stage methods [32].”
Fan, L., Li, Z., Dong, F., et al. (2026) · Infrared Physics & Technology · doi:10.1016/j.infrared.2026.106886
Yang, W., Hou, C., Liu, F., et al. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7387388
Salehian, M., Manjunatha, P., Elefteriadou, L., Kibet, L. J., & Karamouzis, O. (2026) · Journal of Transportation Engineering Part A Systems · doi:10.1061/jtepbs.teeng-9704
Xiao, B., Yang, T., Xie, X.-Q., et al. (2026) · Signal processing. Image communication · doi:10.1016/j.image.2026.117702
Xu, X., He, Z., Lu, J., Ye, Y., Feng, T., & Pi, X. (2026) · Electric Power Systems Research · doi:10.1016/j.epsr.2026.114058
Research on Thrips Object Detection and Counting Algorithms in Leaf Backgrounds
Huang, D., Lei, M., Zhang, Z., et al. (2026) · Agronomy · doi:10.3390/agronomy16171709
Fairness, Bias, and Ethical Governance in Vision Applications
Gupta, B. P. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-9285-1.ch008
Wu, N., Shen, L., Chen, X., et al. (2026) · Applied Sciences · doi:10.3390/app16178780
“Commonly used object detection methods mainly include two-stage detection meth-ods, one-stage detection methods, and Transformer-based detection methods [3].”
Wan, Y., Li, J., Lin, L., Lv, Z., Zhang, Z., Yuan, Q., & Shen, H. (2026) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2026.08.041
Deformable Attention-based Model for Explainable Small Leak Detection in Oil PipelinesPreprint
Zhou, T., Chu, S., 袁易, Gan, Y., Xu, W., & Zheng, W. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7411575
Liu, T.-X., You, M.-W., Zhang, Z.-Y., Sun, Y., Liu, C.-Y., Qian, K., & Lu, X.-Y. (2026) · Sensors · doi:10.3390/s26175655
“…To characterize the offline detection capability of the trained FP32 model before edge-model conversion, the model was evaluated on a desktop GPU using the held-out validation set and standard object-detection metrics, including Precision, Recall, F1-score, and mean Average Precision (mAP) [34,35].”
Quantifying Cross-Illumination Domain Shift in YOLO-Based Camera-Trap Wildlife Detection
Saraboji, K., & Mitra, D. (2026) · IEEE International Conference on Electro Information Technology · doi:10.1109/eit68936.2026.11670345
“Domain shift is a well-established limitation in object detection, where models degrade when evaluated on data distributions different from training [7], [8].”
InfluentialMulti-stage Human Detection Using YOLO to Prompt SAM2
Oliveira, D., Mota, K. O. S., Barros, T., Garrote, L., Premebida, C., & Nunes, U. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-30156-7_10
Safety Detection of Oil and Gas Pipeline Defects Based on Improved YOLOv5
He, Z., Xu, Z., & Li, S. (2026) · International Conference on Advanced Computational Intelligence (ICACI) · doi:10.1109/icaci69731.2026.11677957
Confusion-Pair Loss for Mitigating Structural Confusion in Military Aircraft Detection
Woo, S., & Kim, I. (2026) · IEEE Access · doi:10.1109/access.2026.3732303
“…Faster R-CNN and the YOLO family, deep learning-based object detection The associate editor coordinating the review of this manuscript and approving it for publication was Tao Huang . has achieved substantial gains in mean average precision (mAP) and has become a core enabling technology [1], [2].”
Omar, M., Kamsani, N. A., Sidek, R. M., et al. (2026) · Journal of Intelligent Media Computing · doi:10.47836/jimc.1.1.02
Sardou, A., Rouigueb, A., Demim, F., Amamra, A., & Belaıdı, H. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-26031-4_12
Imad, M., & Shin, S. Y. (2026) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2026.09.006 · Cited by 1
Xu, W., Hu, W., Wang, Y., Zhang, X., Pandey, V., & Qiu, T. (2026) · Chinese Journal of Chemical Engineering · doi:10.1016/j.cjche.2026.08.023
Liu, H., Cao, Z., Ma, Z., et al. (2026) · Foods · doi:10.3390/foods15183211
“These methods first extract candidate target regions and then perform classification and regression localization based on region features [12].”
Enhanced Context-Aware Lightweight YOLO for Accurate Pear Detection in Complex Orchards
Yin, W., Zhang, M., Pan, Z., et al. (2026) · Applied Fruit Science · doi:10.1007/s10341-026-02018-5
“Two-stage detectors offer higher precision at the cost of slower processing, while one-stage detectors are characterized by fast inference speed but relatively lower accuracy (Zou et al. 2023).”
Patch-based CNN for real-time detection of still water in outdoor environments
Sandhu, M. A., Amin, A., Khan, K. B., & Deng, Y. (2026) · Systems and Soft Computing · doi:10.1016/j.sasc.2026.200612
Xia, L. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-38401-0_50
Shape-Adaptive Multi-Scale Feature Enhancement Network for Object Detection in Underwater Vehicle
Zhou, H., & Zhao, X. (2026) · International Conference on Intelligent Systems and Robotics (CISR) · doi:10.1109/cisr70722.2026.11680235
“Object detection is essential to many practical scenarios, including autonomous systems, intelligent surveillance, robotic perception, and environmental monitoring [1].”
Liu, Z., Zhang, Y., Yang, Y., Wen, B., Sha, Y., & Lin, Y. (2026) · Aquaculture International · doi:10.1007/s10499-026-02666-0
“Owing to their strong feature learning capability, detection efficiency, and task adaptability, deep learning methods have become an important technical approach for automatic object detection in complex scenarios (Zou et al. 2023).”
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.