Inner-IoU: More Effective Intersection over Union Loss with Auxiliary Bounding Box

• URL: http://arxiv.org/abs/2311.02877v4
• Date: Tue, 14 Nov 2023 09:49:38 GMT
• Title: Inner-IoU: More Effective Intersection over Union Loss with Auxiliary Bounding Box
• Authors: Hao Zhang, Cong Xu, Shuaijie Zhang
• Abstract summary: We propose Inner-IoU loss, which calculates IoU loss through auxiliary bounding boxes. For different datasets and detectors, we introduce a scaling factor ratio to control the scale size of the auxiliary bounding boxes. Finally, integrate Inner-IoU into the existing IoU-based loss functions for simulation and comparative experiments.
• Score: 10.03001043843768
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: With the rapid development of detectors, Bounding Box Regression (BBR) loss function has constantly updated and optimized. However, the existing IoU-based BBR still focus on accelerating convergence by adding new loss terms, ignoring the limitations of IoU loss term itself. Although theoretically IoU loss can effectively describe the state of bounding box regression,in practical applications, it cannot adjust itself according to different detectors and detection tasks, and does not have strong generalization. Based on the above, we first analyzed the BBR model and concluded that distinguishing different regression samples and using different scales of auxiliary bounding boxes to calculate losses can effectively accelerate the bounding box regression process. For high IoU samples, using smaller auxiliary bounding boxes to calculate losses can accelerate convergence, while larger auxiliary bounding boxes are suitable for low IoU samples. Then, we propose Inner-IoU loss, which calculates IoU loss through auxiliary bounding boxes. For different datasets and detectors, we introduce a scaling factor ratio to control the scale size of the auxiliary bounding boxes for calculating losses. Finally, integrate Inner-IoU into the existing IoU-based loss functions for simulation and comparative experiments. The experiment result demonstrate a further enhancement in detection performance with the utilization of the method proposed in this paper, verifying the effectiveness and generalization ability of Inner-IoU loss. Code is available at https://github.com/malagoutou/Inner-IoU.

Related papers

• Efficient Diffusion as Low Light Enhancer [63.789138528062225]
  Reflectance-Aware Trajectory Refinement (RATR) is a simple yet effective module to refine the teacher trajectory using the reflectance component of images. textbfReflectance-aware textbfDiffusion with textbfDistilled textbfTrajectory (textbfReDDiT) is an efficient and flexible distillation framework tailored for Low-Light Image Enhancement (LLIE)
  arXiv  Detail & Related papers  (2024-10-16T08:07:18Z)
• Unified-IoU: For High-Quality Object Detection [1.62877896907106]
  We propose a new IoU loss function, called Unified-IoU (UIoU), which is more concerned with the weight assignment between different quality prediction boxes. Our proposed method achieves better performance on multiple datasets, especially at a high IoU threshold.
  arXiv  Detail & Related papers  (2024-08-13T04:56:45Z)
• Directly Optimizing IoU for Bounding Box Localization [5.018156030818881]
  This paper presents a novel method to maximize the detection of bounding boxes for the bounding boxes. The Smooth IoU method has shown performance gains over the standard Huber loss. It has been evaluated on the Oxford IIIT, Udacity self-driving car, PA Pets Union, and VWFS Car Damage datasets.
  arXiv  Detail & Related papers  (2023-04-14T17:08:12Z)
• Rethinking IoU-based Optimization for Single-stage 3D Object Detection [103.83141677242871]
  We propose a Rotation-Decoupled IoU (RDIoU) method that can mitigate the rotation-sensitivity issue. Our RDIoU simplifies the complex interactions of regression parameters by decoupling the rotation variable as an independent term.
  arXiv  Detail & Related papers  (2022-07-19T15:35:23Z)
• SIoU Loss: More Powerful Learning for Bounding Box Regression [0.0]
  Loss function SIoU was suggested, where penalty metrics were redefined considering the angle of the vector between the desired regression. Applied to conventional Neural Networks and datasets it is shown that SIoU improves both the speed of training and the accuracy of the inference.
  arXiv  Detail & Related papers  (2022-05-25T12:46:21Z)
• The KFIoU Loss for Rotated Object Detection [115.334070064346]
  In this paper, we argue that one effective alternative is to devise an approximate loss who can achieve trend-level alignment with SkewIoU loss. Specifically, we model the objects as Gaussian distribution and adopt Kalman filter to inherently mimic the mechanism of SkewIoU. The resulting new loss called KFIoU is easier to implement and works better compared with exact SkewIoU.
  arXiv  Detail & Related papers  (2022-01-29T10:54:57Z)
• Alpha-IoU: A Family of Power Intersection over Union Losses for Bounding Box Regression [59.72580239998315]
  We generalize existing IoU-based losses to a new family of power IoU losses that have a power IoU term and an additional power regularization term. Experiments on multiple object detection benchmarks and models demonstrate that $alpha$-IoU losses can surpass existing IoU-based losses by a noticeable performance margin.
  arXiv  Detail & Related papers  (2021-10-26T13:09:20Z)
• SCALoss: Side and Corner Aligned Loss for Bounding Box Regression [29.275260127860783]
  We propose Side Overlap (SO) loss by maximizing the side overlap of two bounding boxes, which puts more penalty for low overlapping bounding box cases. To speed up the convergence, the Corner Distance (CD) is added into the objective function. We get a new regression objective function, Side and Corner Align Loss (SCALoss)
  arXiv  Detail & Related papers  (2021-04-01T13:46:35Z)
• Focal and Efficient IOU Loss for Accurate Bounding Box Regression [63.14659624634066]
  In object detection, bounding box regression (BBR) is a crucial step that determines the object localization performance. Most previous loss functions for BBR have two main drawbacks: (i) Both $ell_n$-norm and IOU-based loss functions are inefficient to depict the objective of BBR, which leads to slow convergence and inaccurate regression results.
  arXiv  Detail & Related papers  (2021-01-20T14:33:58Z)
• AQD: Towards Accurate Fully-Quantized Object Detection [94.06347866374927]
  We propose an Accurate Quantized object Detection solution, termed AQD, to get rid of floating-point computation. Our AQD achieves comparable or even better performance compared with the full-precision counterpart under extremely low-bit schemes.
  arXiv  Detail & Related papers  (2020-07-14T09:07:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.