Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Arc Length Function01:22

Arc Length Function

94
The arc length function represents the total distance traveled along a smooth curve measured from a fixed starting point to a variable endpoint. For curves that are continuous and differentiable, arc length provides a precise way to quantify distance when straight-line approximations are insufficient.To derive arc length, the curve is divided into many small segments. Each segment is approximated by a straight line whose length depends on the horizontal and vertical changes over that interval.
94
Radius of Gyration of an Area01:12

Radius of Gyration of an Area

3.0K
The second moment of area, also known as the moment of inertia of area, is a crucial factor in understanding an object's resistance against bending deformation, or stiffness. To accurately estimate the second moment of area along any axis, one needs to concentrate all areas associated with that object into a thin strip, which should be placed parallel to that particular axis.
3.0K
Arc Length of a Curve: Problem Solving01:21

Arc Length of a Curve: Problem Solving

101
A high-voltage power line spans a 40-meter horizontal distance between two transmission towers, resulting in a 10-meter vertical sag due to the effects of gravity and thermal expansion. The curve formed by the suspended cable is a catenary, which accurately models the behavior of a uniform, flexible cable under its own weight. Unlike a parabolic shape, the catenary is described by the hyperbolic cosine function and offers a precise representation of the cable's form.In this setup, engineers...
101
Mean Absolute Deviation01:13

Mean Absolute Deviation

3.6K
The mean absolute deviation is also a measure of the variability of data in a sample. It is the absolute value of the average difference between the data values and the mean.
Let us consider a dataset containing the number of unsold cupcakes in five shops: 10, 15, 8, 7, and 10. Initially, calculate the sample mean. Then calculate the deviation, or the difference, between each data value and the mean. Next, the absolute values of these deviations are added and divided by the sample size to...
3.6K
Distance Problem01:29

Distance Problem

130
When an object's velocity changes over time, the total distance traveled can be determined by summing small displacement intervals over short increments. This approach approximates the true distance through numerical summation and the use of integral calculus. An estimate of the total displacement can be obtained by measuring velocity at regular intervals and multiplying each value by the corresponding time step.If a runner accelerates over the first three seconds of a race, speed measurements...
130
Linear Approximation in Frequency Domain01:26

Linear Approximation in Frequency Domain

414
Linear systems are characterized by two main properties: superposition and homogeneity. Superposition allows the response to multiple inputs to be the sum of the responses to each individual input. Homogeneity ensures that scaling an input by a scalar results in the response being scaled by the same scalar.
In contrast, nonlinear systems do not inherently possess these properties. However, for small deviations around an operating point, a nonlinear system can often be approximated as linear....
414

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Rhodium(III)-Catalyzed [3 + 2] Annulation of Diazo Compounds with Enaminones for Constructing Polysubstituted NH-Pyrroles.

Organic letters·2026
Same author

Divergent Three-Component Assembly of Densely Functionalized Dibenzofurans via ZnCl<sub>2</sub>-Mediated Cascade Annulation.

Organic letters·2026
Same author

Nonlinear Modeling and Differential-Voltage Control of an Electrostatic MEMS Micromirror for Miniaturized Laser Communication Terminals.

Micromachines·2026
Same author

Antifungal efficacy of a novel type of nanosalt against human fungal pathogens and its antifungal mechanisms.

Applied microbiology and biotechnology·2026
Same author

A network meta-analysis of the performance of acupoint stimulation therapy in improving fatigue, neurological function, and activities of daily living in patients with multiple sclerosis.

Frontiers in neurology·2026
Same author

DNA cube-gated ratiometric ECL/SERS biosensor for APE1 via HCR-driven interfacial reconstruction.

Biosensors & bioelectronics·2026

Related Experiment Video

Updated: Mar 15, 2026

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.7K

ATDIOU: Arctangent Differential Loss Function for Bounding Box Regression.

Qiang Tang1,2, Hao Qiang1,2, Yuan Tian1,2

  • 1Xi'an Institute of Optics and Precision Mechanics of CAS, Xi'an 710119, China.

Sensors (Basel, Switzerland)
|March 14, 2026
PubMed
Summary

We introduce ATDIoU, a new arctangent-differential loss for bounding box regression. This method improves object detection accuracy by reducing sensitivity to localization errors, outperforming existing approaches.

Keywords:
arctangent-differential functionbounding box regressioncomputer visionobject detection

More Related Videos

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing
06:25

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Published on: February 23, 2024

1.2K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.8K

Related Experiment Videos

Last Updated: Mar 15, 2026

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine
07:05

Visualization Method for Proprioceptive Drift on a 2D Plane Using Support Vector Machine

Published on: October 27, 2016

9.7K
Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing
06:25

Author Spotlight: Assessment of Visual Acuity in Central Vision Loss Through Motion-Based Peripheral Vision Testing

Published on: February 23, 2024

1.2K
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
13:44

Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns

Published on: August 30, 2013

43.8K

Area of Science:

  • Computer Vision
  • Machine Learning

Background:

  • Object detection is crucial in computer vision.
  • Bounding box regression (BBR) losses significantly impact detector performance.
  • Intersection over Union (IoU) based metrics are sensitive to positional deviations, hindering optimization.

Purpose of the Study:

  • To propose ATDIoU, a novel arctangent-differential loss for bounding-box regression.
  • To mitigate bounding box drift and reduce sensitivity to localization errors in object detection.
  • To enhance the effectiveness of models in learning target positions.

Main Methods:

  • Developed ATDIoU, a novel loss function for BBR.
  • Modeled distances between predicted and ground truth box vertices using a 2D arctangent differential distribution (ATD).
  • Integrated ATDIoU into the YOLOv6 object detection framework.

Main Results:

  • ATDIoU demonstrated improved performance in object detection tasks.
  • Experiments conducted on PASCAL VOC and VisDrone2019 datasets.
  • Achieved 1.4% and 0.7% mean average precision (mAP) gains over MPDIoU on respective datasets.

Conclusions:

  • ATDIoU effectively mitigates bounding box drift and localization errors.
  • The proposed loss function guides models to learn target positions more accurately.
  • ATDIoU offers a promising advancement for bounding box regression in object detection.