TSAA: A Two-Stage Anchor Assignment Method towards Anchor Drift in Crowded Object Detection

• URL: http://arxiv.org/abs/2211.00826v1
• Date: Wed, 2 Nov 2022 02:05:00 GMT
• Title: TSAA: A Two-Stage Anchor Assignment Method towards Anchor Drift in Crowded Object Detection
• Abstract summary: A positive anchor does not always regress toward the object that overlaps it the most when multiple objects overlap. A simple but efficient adaptive two-stage anchor assignment (TSAA) method is proposed.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Among current anchor-based detectors, a positive anchor box will be intuitively assigned to the object that overlaps it the most. The assigned label to each anchor will directly determine the optimization direction of the corresponding prediction box, including the direction of box regression and category prediction. In our practice of crowded object detection, however, the results show that a positive anchor does not always regress toward the object that overlaps it the most when multiple objects overlap. We name it anchor drift. The anchor drift reflects that the anchor-object matching relation, which is determined by the degree of overlap between anchors and objects, is not always optimal. Conflicts between the fixed matching relation and learned experience in the past training process may cause ambiguous predictions and thus raise the false-positive rate. In this paper, a simple but efficient adaptive two-stage anchor assignment (TSAA) method is proposed. It utilizes the final prediction boxes rather than the fixed anchors to calculate the overlap degree with objects to determine which object to regress for each anchor. The participation of the prediction box makes the anchor-object assignment mechanism adaptive. Extensive experiments are conducted on three classic detectors RetinaNet, Faster-RCNN and YOLOv3 on CrowdHuman and COCO to evaluate the effectiveness of TSAA. The results show that TSAA can significantly improve the detectors' performance without additional computational costs or network structure changes.

Related papers

• Domain Adaptive Object Detection via Dual-Stream Bilevel-Cycle Optimization [64.83565413108789]
  Cycle self-training (CST) breaks the shared classifier assumption of the standard self-training framework.<n>CST exploits unlabeled target data by training with target pseudo-labels.
  arXiv  Detail & Related papers  (2026-06-30T09:05:30Z)
• IoUCert: Robustness Verification for Anchor-based Object Detectors [58.35703549470485]
  We introduce IoUCert, a novel formal verification framework designed specifically to overcome these bottlenecks in anchor-based object detection architectures.<n>We show that our method enables the robustness verification of realistic, anchor-based models including SSD, YOLOv2, and YOLOv3 variants against various input perturbations.
  arXiv  Detail & Related papers  (2026-03-03T14:36:46Z)
• PointOBB: Learning Oriented Object Detection via Single Point Supervision [55.88982271340328]
  This paper proposes PointOBB, the first single Point-based OBB generation method, for oriented object detection. PointOBB operates through the collaborative utilization of three distinctive views: an original view, a resized view, and a rotated/flipped (rot/flp) view. Experimental results on the DIOR-R and DOTA-v1.0 datasets demonstrate that PointOBB achieves promising performance.
  arXiv  Detail & Related papers  (2023-11-23T15:51:50Z)
• Semi-DETR: Semi-Supervised Object Detection with Detection Transformers [105.45018934087076]
  We analyze the DETR-based framework on semi-supervised object detection (SSOD) We present Semi-DETR, the first transformer-based end-to-end semi-supervised object detector. Our method outperforms all state-of-the-art methods by clear margins.
  arXiv  Detail & Related papers  (2023-07-16T16:32:14Z)
• SOOD: Towards Semi-Supervised Oriented Object Detection [57.05141794402972]
  This paper proposes a novel Semi-supervised Oriented Object Detection model, termed SOOD, built upon the mainstream pseudo-labeling framework. Our experiments show that when trained with the two proposed losses, SOOD surpasses the state-of-the-art SSOD methods under various settings on the DOTA-v1.5 benchmark.
  arXiv  Detail & Related papers  (2023-04-10T11:10:42Z)
• SAILOR: Scaling Anchors via Insights into Latent Object Representation [5.96002531660335]
  LiDAR 3D object detection models are inevitably biased towards their training dataset. object sizes vary heavily between domains due to, for instance, different labeling policies or geographical locations. We propose an unsupervised anchor calibration method to overcome the object size bias.
  arXiv  Detail & Related papers  (2022-10-14T13:40:46Z)
• Decoupled Adaptation for Cross-Domain Object Detection [69.5852335091519]
  Cross-domain object detection is more challenging than object classification. D-adapt achieves a state-of-the-art results on four cross-domain object detection tasks.
  arXiv  Detail & Related papers  (2021-10-06T08:43:59Z)
• Pseudo-IoU: Improving Label Assignment in Anchor-Free Object Detection [60.522877583407904]
  Current anchor-free object detectors are quite simple and effective yet lack accurate label assignment methods. We present Pseudo-Intersection-over-Union(Pseudo-IoU): a simple metric that brings more standardized and accurate assignment rule into anchor-free object detection frameworks. Our method achieves comparable performance to other recent state-of-the-art anchor-free methods without bells and whistles.
  arXiv  Detail & Related papers  (2021-04-29T02:48:47Z)
• Dynamic Anchor Learning for Arbitrary-Oriented Object Detection [4.247967690041766]
  Arbitrary-oriented objects widely appear in natural scenes, aerial photographs, remote sensing images, etc. Current rotation detectors use plenty of anchors with different orientations to achieve spatial alignment with ground truth boxes. We propose a dynamic anchor learning (DAL) method, which utilizes the newly defined matching degree.
  arXiv  Detail & Related papers  (2020-12-08T01:30:06Z)
• Align Deep Features for Oriented Object Detection [40.28244152216309]
  We propose a single-shot Alignment Network (S$2$A-Net) consisting of two modules: a Feature Alignment Module (FAM) and an Oriented Detection Module (ODM) The FAM can generate high-quality anchors with an Anchor Refinement Network and adaptively align the convolutional features according to the anchor boxes with a novel Alignment Convolution. The ODM first adopts active rotating filters to encode the orientation information and then produces orientation-sensitive and orientation-invariant features to alleviate the inconsistency between classification score and localization accuracy.
  arXiv  Detail & Related papers  (2020-08-21T09:55:13Z)
• Probabilistic Anchor Assignment with IoU Prediction for Object Detection [9.703212439661097]
  In object detection, determining which anchors to assign as positive or negative samples, known as anchor assignment, has been revealed as a core procedure that can significantly affect a model's performance. We propose a novel anchor assignment strategy that adaptively separates anchors into positive and negative samples for a ground truth bounding box according to the model's learning status.
  arXiv  Detail & Related papers  (2020-07-16T04:26:57Z)
• Scope Head for Accurate Localization in Object Detection [135.9979405835606]
  We propose a novel detector coined as ScopeNet, which models anchors of each location as a mutually dependent relationship. With our concise and effective design, the proposed ScopeNet achieves state-of-the-art results on COCO.
  arXiv  Detail & Related papers  (2020-05-11T04:00:09Z)

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.