Object detection is a computer technology related tocomputer vision andimage processing that deals with detecting instances of semantic objects of a certain class (such as humans, buildings, or cars) in digital images and videos.^[1] Well-researched domains of object detection includeface detection andpedestrian detection. Object detection has applications in many areas of computer vision, includingimage retrieval andvideo surveillance.

It is widely used incomputer vision tasks such asimage annotation,^[2] vehicle counting,^[3]activity recognition,^[4]face detection,face recognition,video object co-segmentation. It is also used intracking objects, for example tracking a ball during a football match, tracking movement of a cricket bat, or tracking a person in a video.

Often, the test images are sampled from a different data distribution, making the object detection task significantly more difficult.^[5] To address the challenges caused by the domain gap between training and test data, many unsupervised domain adaptation approaches have been proposed.^{[5][6][7][8][9]} A simple and straightforward solution for reducing the domain gap is to apply an image-to-image translation approach, such as cycle-GAN.^[10] Among other uses, cross-domain object detection is applied in autonomous driving, where models can be trained on a vast amount of video game scenes, since the labels can be generated without manual labor.

Every object class has its own specialfeatures that help in classifying the class – for example allcircles are round.Object class detection uses these special features. For example, when looking for circles, objects that are at a particular distance from a point (i.e. the center) are sought. Similarly, when looking for squares, objects that areperpendicular at corners and have equal side lengths are needed. A similar approach is used forface identification where eyes, nose, and lips can be found andfeatures like skin color and distance between eyes can be found.

Intersection over union as a similarity measure for object detection on images – an important task incomputer vision

For object localization, true positive is often measured by the thresholdedintersection over union. For example, if there is a traffic sign in the image, with a bounding box drawn by a human ("ground truth label"), then a neural network has detected the traffic sign (atrue positive) at 0.5 threshold iff it has drawn a bounding box whose IoU with the ground truth is above 0.5. Otherwise, the bounding box is afalse positive.

If there is only a single ground truth bounding box, but multiple predictions, then the IoU of each prediction is calculated. The prediction with the highest IoU is a true positive if it is above threshold, else it is a false positive. All other predicted bounding boxes are false positives. If there is no prediction with an IoU above the threshold, then the ground truth label has afalse negative.

For simultaneous object localization and classification, a true positive is one where the class label is correct, and the bounding box has an IoU exceeding the threshold.

Simultaneous object localization and classification is benchmarked by themean average precision (mAP). The average precision (AP) of the network for a class of objects is the area under theprecision-recall curve as the confidence threshold is varied. The mAP is the average of AP over all classes.

Simplified example of training a neural network in object detection: The network is trained by multiple images that are known to depictstarfish andsea urchins, which are correlated with "nodes" that represent visualfeatures. The starfish match with a ringed texture and a star outline, whereas most sea urchins match with a striped texture and oval shape. However, the instance of a ring textured sea urchin creates a weakly weighted association between them.

Subsequent run of the network on an input image (left):^[11] The network correctly detects the starfish. However, the weakly weighted association between ringed texture and sea urchin also confers a weak signal to the latter from one of two intermediate nodes. In addition, a shell that was not included in the training gives a weak signal for the oval shape, also resulting in a weak signal for the sea urchin output. These weak signals may result in afalse positive result for sea urchin.
In reality, textures and outlines would not be represented by single nodes, but rather by associated weight patterns of multiple nodes.

Methods for object detection generally fall into either neural network-based or non-neural approaches. For non-neural approaches, it becomes necessary to first define features using one of the methods below, then using a technique such assupport vector machine (SVM) to do the classification. On the other hand, neural techniques are able to do end-to-end object detection without specifically defining features, and are typically based onconvolutional neural networks (CNN).

• Non-neural approaches:
  • Viola–Jones object detection framework based onHaar features
  • Scale-invariant feature transform (SIFT)
  • Histogram of oriented gradients (HOG) features^[12]
• Neural network approaches:
  • OverFeat.^[13]
  • Region Proposals (R-CNN,^[14] Fast R-CNN,^[15] Faster R-CNN,^[16] cascade R-CNN.^[17])
  • You Only Look Once (YOLO).^[18]
  • Single Shot MultiBox Detector (SSD)^[19]
  • Single-Shot Refinement Neural Network for Object Detection (RefineDet)^[20]
  • Retina-Net^[21][17]
  • Deformable convolutional networks^[22][23]

• Feature detection (computer vision)
• Moving object detection
• Small object detection
• Outline of object recognition
• Teknomo–Fernandez algorithm

• Zou, Zhengxia; Chen, Keyan; Shi, Zhenwei; Guo, Yuhong; Ye, Jieping (March 2023). "Object Detection in 20 Years: A Survey".Proceedings of the IEEE.111 (3):257–276.doi:10.1109/JPROC.2023.3238524.ISSN 0018-9219.

• Joshi, Snehal (2023-10-24)."Top Object Detection Models".hitechbpo.com.
• Weng, Lilian (2017-10-29)."Object Detection for Dummies Part 1: Gradient Vector, HOG, and SS".lilianweng.github.io. Retrieved2024-09-11.
• Weng, Lilian (2017-12-15)."Object Detection for Dummies Part 2: CNN, DPM and Overfeat".lilianweng.github.io. Retrieved2024-09-11.
• Weng, Lilian (2017-12-31)."Object Detection for Dummies Part 3: R-CNN Family".lilianweng.github.io. Retrieved2024-09-11.
• Weng, Lilian (2018-12-27)."Object Detection Part 4: Fast Detection Models".lilianweng.github.io. Retrieved2024-09-11.
• Multiple object class detection
• Spatio-temporal action localization
• Online Object Detection Demo
• Video object detection and co-segmentation

• Object recognition and categorization
• Surveillance
• Applications of computer vision
• Gesture recognition

• CS1 maint: multiple names: authors list
• Articles with short description
• Short description matches Wikidata
• Pages using multiple image with auto scaled images