P2P-Loc: Point to Point Tiny Person Localization

• URL: http://arxiv.org/abs/2112.15344v1
• Date: Fri, 31 Dec 2021 08:24:43 GMT
• Title: P2P-Loc: Point to Point Tiny Person Localization
• Authors: Xuehui Yu, Di Wu, Qixiang Ye, Jianbin Jiao and Zhenjun Han
• Abstract summary: We propose a novel point-based framework for the person localization task. Annotating each person as a coarse point (CoarsePoint) can be any point within the object extent, instead of an accurate bounding box. Our approach achieves comparable object localization performance while saving annotation cost up to 80$%$.
• Score: 47.6728595874315
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Bounding-box annotation form has been the most frequently used method for visual object localization tasks. However, bounding-box annotation relies on the large amounts of precisely annotating bounding boxes, which is expensive, laborious, thus impossible in practical scenarios, and even redundant for some applications caring not about size. Therefore, we propose a novel point-based framework for the person localization task by annotating each person as a coarse point (CoarsePoint) which can be any point within the object extent, instead of an accurate bounding box. And then predict the person's location as a 2D coordinate in the image. That greatly simplifies the data annotation pipeline. However, the CoarsePoint annotation inevitably causes the label reliability decrease (label uncertainty) and network confusion during training. As a result, we propose a point self-refinement approach, which iteratively updates point annotations in a self-paced way. The proposed refinement system alleviates the label uncertainty and progressively improves localization performance. Experiments show that our approach achieves comparable object localization performance while saving annotation cost up to 80$\%$. Code is enclosed in the supplementary materials.

Related papers

• Dense Center-Direction Regression for Object Counting and Localization with Point Supervision [1.9526430269580954]
  We propose a novel approach termed CeDiRNet for point-supervised learning. It uses a dense regression of directions pointing towards the nearest object centers. We show that it outperforms the existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-08-26T17:49:27Z)
• CPR++: Object Localization via Single Coarse Point Supervision [55.8671776333499]
  coarse point refinement (CPR) is first attempt to alleviate semantic variance from an algorithmic perspective. CPR reduces semantic variance by selecting a semantic centre point in a neighbourhood region to replace the initial annotated point. CPR++ can obtain scale information and further reduce the semantic variance in a global region.
  arXiv  Detail & Related papers  (2024-01-30T17:38:48Z)
• Knowledge Combination to Learn Rotated Detection Without Rotated Annotation [53.439096583978504]
  Rotated bounding boxes drastically reduce output ambiguity of elongated objects. Despite the effectiveness, rotated detectors are not widely employed. We propose a framework that allows the model to predict precise rotated boxes.
  arXiv  Detail & Related papers  (2023-04-05T03:07:36Z)
• Few-shot Geometry-Aware Keypoint Localization [13.51645400661565]
  We present a novel formulation that learns to localize semantically consistent keypoint definitions. We use a few user-labeled 2D images as input examples, which are extended via self-supervision. We introduce 3D geometry-aware constraints to uplift keypoints, achieving more accurate 2D localization.
  arXiv  Detail & Related papers  (2023-03-30T08:19:42Z)
• Point-Teaching: Weakly Semi-Supervised Object Detection with Point Annotations [81.02347863372364]
  We present Point-Teaching, a weakly semi-supervised object detection framework. Specifically, we propose a Hungarian-based point matching method to generate pseudo labels for point annotated images. We propose a simple-yet-effective data augmentation, termed point-guided copy-paste, to reduce the impact of the unmatched points.
  arXiv  Detail & Related papers  (2022-06-01T07:04:38Z)
• Rethinking Localization Map: Towards Accurate Object Perception with Self-Enhancement Maps [78.2581910688094]
  This work introduces a novel self-enhancement method to harvest accurate object localization maps and object boundaries with only category labels as supervision. In particular, the proposed Self-Enhancement Maps achieve the state-of-the-art localization accuracy of 54.88% on ILSVRC.
  arXiv  Detail & Related papers  (2020-06-09T12:35:55Z)
• Weakly-Supervised Salient Object Detection via Scribble Annotations [54.40518383782725]
  We propose a weakly-supervised salient object detection model to learn saliency from scribble labels. We present a new metric, termed saliency structure measure, to measure the structure alignment of the predicted saliency maps. Our method not only outperforms existing weakly-supervised/unsupervised methods, but also is on par with several fully-supervised state-of-the-art models.
  arXiv  Detail & Related papers  (2020-03-17T12:59:50Z)
• Rethinking the Route Towards Weakly Supervised Object Localization [28.90792512056726]
  We show that weakly supervised object localization should be divided into two parts: class-agnostic object localization and object classification. For class-agnostic object localization, we should use class-agnostic methods to generate noisy pseudo annotations and then perform bounding box regression on them without class labels. Our PSOL models have good transferability across different datasets without fine-tuning.
  arXiv  Detail & Related papers  (2020-02-26T08:54:20Z)

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.