Learning Where to Embed: Noise-Aware Positional Embedding for Query Retrieval in Small-Object Detection

• URL: http://arxiv.org/abs/2604.15065v1
• Date: Thu, 16 Apr 2026 14:28:36 GMT
• Title: Learning Where to Embed: Noise-Aware Positional Embedding for Query Retrieval in Small-Object Detection
• Abstract summary: Transformer-based detectors have advanced small-object detection, but they often remain inefficient and vulnerable to background-induced query noise.<n>We present HELP, a noise-aware positional-semantic fusion framework that studies where to embed positional information.<n>Within HELP, we introduce Heatmap-guided Positional Embedding (HPE) as the core embedding mechanism and visualize it with a heatbar for interpretable diagnosis and fine-tuning.
• Score: 2.2251429968996947
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Transformer-based detectors have advanced small-object detection, but they often remain inefficient and vulnerable to background-induced query noise, which motivates deep decoders to refine low-quality queries. We present HELP (Heatmap-guided Embedding Learning Paradigm), a noise-aware positional-semantic fusion framework that studies where to embed positional information by selectively preserving positional encodings in foreground-salient regions while suppressing background clutter. Within HELP, we introduce Heatmap-guided Positional Embedding (HPE) as the core embedding mechanism and visualize it with a heatbar for interpretable diagnosis and fine-tuning. HPE is integrated into both the encoder and decoder: it guides noise-suppressed feature encoding by injecting heatmap-aware positional encoding, and it enables high-quality query retrieval by filtering background-dominant embeddings via a gradient-based mask filter before decoding. To address feature sparsity in complex small targets, we integrate Linear-Snake Convolution to enrich retrieval-relevant representations. The gradient-based heatmap supervision is used during training only, incurring no additional gradient computation at inference. As a result, our design reduces decoder layers from eight to three and achieves a 59.4% parameter reduction (66.3M vs. 163M) while maintaining consistent accuracy gains under a reduced compute budget across benchmarks. Code Repository: https://github.com/yidimopozhibai/Noise-Suppressed-Query-Retrieval

Related papers

• HGSQ: Heatmap-Guided Sparse Query Detector for Real-Time Aerial Small Object Detection [0.0]
  Real-time aerial small object detection requires a detector to preserve fine-grained localization while avoiding redundant computation on large background regions.<n>This paper proposes HGSQ, a Heatmap-Guided Sparse Query Detector for real-time aerial small object detection.<n>HGSQ treats the heatmap as a real-time computation budget rather than a visualization map during deployment.
  arXiv  Detail & Related papers  (2026-09-10T12:52:34Z)
• Generative MIMO Beam Map Construction for Location Recovery and Beam Tracking [67.65578956523403]
  This paper proposes a generative framework to recover location labels directly from sparse channel state information (CSI) measurements.<n>Instead of directly storing raw CSI, we learn a compact low-dimensional radio map embedding and leverage a generative model to reconstruct the high-dimensional CSI.<n> Numerical experiments demonstrate that the proposed model can improve localization accuracy by over 30% and achieve a 20% capacity gain in non-line-of-sight (NLOS) scenarios.
  arXiv  Detail & Related papers  (2025-11-21T07:25:49Z)
• DoPE: Denoising Rotary Position Embedding [60.779039511252584]
  Rotary Position Embedding (RoPE) in Transformer models has inherent limits that weaken length.<n>We reinterpret the attention map with positional encoding as a noisy feature map, and propose Denoising Positional extrapolation page (DoPE)<n>DoPE is a training-free method based on truncated matrix entropy to detect outlier frequency bands in the feature map.
  arXiv  Detail & Related papers  (2025-11-12T09:32:35Z)
• HMPE:HeatMap Embedding for Efficient Transformer-Based Small Object Detection [0.0]
  This paper introduces HeatMap Position Embedding (HMPE), a novel Transformer Optimization technique.<n>HMPE dynamically integrates positional encoding with semantic detection information through heatmap-guided adaptive learning.<n>By employing HMPE-enhanced embedding, we are able to reduce the number of decoder layers from eight to a minimum of three, significantly decreasing both inference and training costs.
  arXiv  Detail & Related papers  (2025-04-18T05:24:08Z)
• A new baseline for edge detection: Make Encoder-Decoder great again [3.3171122239461193]
  The proposed New Baseline for Edge Detection (NBED) achieves superior performance consistently across multiple edge detection benchmarks. The ODS of NBED on BSDS500 is 0.838, achieving state-of-the-art performance.
  arXiv  Detail & Related papers  (2024-09-23T12:54:38Z)
• Learning to Make Keypoints Sub-Pixel Accurate [80.55676599677824]
  This work addresses the challenge of sub-pixel accuracy in detecting 2D local features. We propose a novel network that enhances any detector with sub-pixel precision by learning an offset vector for detected features.
  arXiv  Detail & Related papers  (2024-07-16T12:39:56Z)
• Deep Homography Estimation for Visual Place Recognition [49.235432979736395]
  We propose a transformer-based deep homography estimation (DHE) network. It takes the dense feature map extracted by a backbone network as input and fits homography for fast and learnable geometric verification. Experiments on benchmark datasets show that our method can outperform several state-of-the-art methods.
  arXiv  Detail & Related papers  (2024-02-25T13:22:17Z)
• V-DETR: DETR with Vertex Relative Position Encoding for 3D Object Detection [73.37781484123536]
  We introduce a highly performant 3D object detector for point clouds using the DETR framework. To address the limitation, we introduce a novel 3D Relative Position (3DV-RPE) method. We show exceptional results on the challenging ScanNetV2 benchmark.
  arXiv  Detail & Related papers  (2023-08-08T17:14:14Z)
• Progressive Coordinate Transforms for Monocular 3D Object Detection [52.00071336733109]
  We propose a novel and lightweight approach, dubbed em Progressive Coordinate Transforms (PCT) to facilitate learning coordinate representations. In this paper, we propose a novel and lightweight approach, dubbed em Progressive Coordinate Transforms (PCT) to facilitate learning coordinate representations.
  arXiv  Detail & Related papers  (2021-08-12T15:22:33Z)
• CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depth [83.77839773394106]
  We present a lightweight, tightly-coupled deep depth network and visual-inertial odometry system. We provide the network with previously marginalized sparse features from VIO to increase the accuracy of initial depth prediction. We show that it can run in real-time with single-thread execution while utilizing GPU acceleration only for the network and code Jacobian.
  arXiv  Detail & Related papers  (2020-12-18T09:42:54Z)

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.