ASAG: Building Strong One-Decoder-Layer Sparse Detectors via Adaptive Sparse Anchor Generation

• URL: http://arxiv.org/abs/2308.09242v1
• Date: Fri, 18 Aug 2023 02:06:49 GMT
• Title: ASAG: Building Strong One-Decoder-Layer Sparse Detectors via Adaptive Sparse Anchor Generation
• Authors: Shenghao Fu, Junkai Yan, Yipeng Gao, Xiaohua Xie, Wei-Shi Zheng
• Abstract summary: We bridge the performance gap between sparse and dense detectors by proposing Adaptive Sparse Anchor Generator (ASAG) ASAG predicts dynamic anchors on patches rather than grids in a sparse way so that it alleviates the feature conflict problem. Our method outperforms dense-d ones and achieves a better speed-accuracy trade-off.
• Score: 50.01244854344167
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent sparse detectors with multiple, e.g. six, decoder layers achieve promising performance but much inference time due to complex heads. Previous works have explored using dense priors as initialization and built one-decoder-layer detectors. Although they gain remarkable acceleration, their performance still lags behind their six-decoder-layer counterparts by a large margin. In this work, we aim to bridge this performance gap while retaining fast speed. We find that the architecture discrepancy between dense and sparse detectors leads to feature conflict, hampering the performance of one-decoder-layer detectors. Thus we propose Adaptive Sparse Anchor Generator (ASAG) which predicts dynamic anchors on patches rather than grids in a sparse way so that it alleviates the feature conflict problem. For each image, ASAG dynamically selects which feature maps and which locations to predict, forming a fully adaptive way to generate image-specific anchors. Further, a simple and effective Query Weighting method eases the training instability from adaptiveness. Extensive experiments show that our method outperforms dense-initialized ones and achieves a better speed-accuracy trade-off. The code is available at \url{https://github.com/iSEE-Laboratory/ASAG}.

Related papers

• Global Context Aggregation Network for Lightweight Saliency Detection of Surface Defects [70.48554424894728]
  We develop a Global Context Aggregation Network (GCANet) for lightweight saliency detection of surface defects on the encoder-decoder structure. First, we introduce a novel transformer encoder on the top layer of the lightweight backbone, which captures global context information through a novel Depth-wise Self-Attention (DSA) module. The experimental results on three public defect datasets demonstrate that the proposed network achieves a better trade-off between accuracy and running efficiency compared with other 17 state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-09-22T06:19:11Z)
• Efficient Person Search: An Anchor-Free Approach [86.45858994806471]
  Person search aims to simultaneously localize and identify a query person from realistic, uncropped images. To achieve this goal, state-of-the-art models typically add a re-id branch upon two-stage detectors like Faster R-CNN. In this work, we present an anchor-free approach to efficiently tackling this challenging task, by introducing the following dedicated designs.
  arXiv  Detail & Related papers  (2021-09-01T07:01:33Z)
• Learning Spatio-Temporal Transformer for Visual Tracking [108.11680070733598]
  We present a new tracking architecture with an encoder-decoder transformer as the key component. The whole method is end-to-end, does not need any postprocessing steps such as cosine window and bounding box smoothing. The proposed tracker achieves state-of-the-art performance on five challenging short-term and long-term benchmarks, while running real-time speed, being 6x faster than Siam R-CNN.
  arXiv  Detail & Related papers  (2021-03-31T15:19:19Z)
• Anchor-Free Person Search [127.88668724345195]
  Person search aims to simultaneously localize and identify a query person from realistic, uncropped images. Most existing works employ two-stage detectors like Faster-RCNN, yielding encouraging accuracy but with high computational overhead. We present the Feature-Aligned Person Search Network (AlignPS), the first anchor-free framework to efficiently tackle this challenging task.
  arXiv  Detail & Related papers  (2021-03-22T07:04:29Z)
• Unsupervised Domain Adaptation from Synthetic to Real Images for Anchorless Object Detection [0.0]
  This paper implements unsupervised domain adaptation methods on an anchorless object detector. In our work, we use CenterNet, one of the most recent anchorless architectures, for a domain adaptation problem involving synthetic images.
  arXiv  Detail & Related papers  (2020-12-15T10:51:43Z)
• 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)
• FCOS: A simple and strong anchor-free object detector [111.87691210818194]
  We propose a fully convolutional one-stage object detector (FCOS) to solve object detection in a per-pixel prediction fashion. Almost all state-of-the-art object detectors such as RetinaNet, SSD, YOLOv3, and Faster R-CNN rely on pre-defined anchor boxes. In contrast, our proposed detector FCOS is anchor box free, as well as proposal free.
  arXiv  Detail & Related papers  (2020-06-14T01:03:39Z)

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.