NGIoU Loss: Generalized Intersection over Union Loss Based on a New Bounding Box Regression

نویسندگان

چکیده

Loss functions, such as the IoU function and GIoU (Generalized Intersection over Union) have been put forward to replace regression loss functions commonly used in calculation. alleviates vanishing gradient case of non-overlapping, but it will completely degenerate into when bounding boxes overlap totally, which fails achieve optimization effect. To solve this problem, some improvements are proposed paper on basis function, taking account rate complete boxes. In PASCAL VOC data, experimental results demonstrate that AP NGIoU YOLOv4 model is 47.68%, 1.15% higher than highest map value 86.79% YOLOv5 model.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Sweep-plane Approach to Bounding Box Intersection

This paper summarises an effort towards approximation of an optimal sweepplane approach to axis aligned bounding box intersection problem. In particular a spatial hash table and priority search tree are combined in order to obtain a data structure suitable for solving two-dimensional dynamic rectangle intersection problem. Some variants of this structure allow logarithmic update and query times...

متن کامل

Superset Learning Based on Generalized Loss Minimization

In standard supervised learning, each training instance is associated with an outcome from a corresponding output space (e.g., a class label in classification or a real number in regression). In the superset learning problem, the outcome is only characterized in terms of a superset—a subset of candidates that covers the true outcome but may also contain additional ones. Thus, superset learning ...

متن کامل

Support Vector Regression with a Generalized Quadratic Loss

The standard SVR formulation for real-valued function approximation on multidimensional spaces is based on the -insensitive loss function, where errors are considered not correlated. Due to this, local information in the feature space which can be useful to improve the prediction model is disregarded. In this paper we address this problem by defining a generalized quadratic loss where the co-oc...

متن کامل

Bounding the Range of a Rational Function over a box

A simple method is presented by which tight bounds on the range of a multivariate rational function over a box can be computed. The approach relies on the expansion of the numerator and denominator polynomials into Bernstein polynomials.

متن کامل

Introducing a New Method for Multiarea Transmission Networks Loss Allocation

Transmission loss allocation in very large networks with multiple interconnected areas or countries is investigated in this paper. The main contribution is to propose a method to calculate the amount of losses due to activity of each participant in the multi area markets. Pricing of cross-border trades in Multi area systems is often difficult since individual countries may use incompatible ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Applied sciences

سال: 2022

ISSN: ['2076-3417']

DOI: https://doi.org/10.3390/app122412785