# SpotNet: Self-Attention Multi-Task Network for Object Detection

arXiv: https://arxiv.org/abs/2002.05540 · HTML: https://mlsanity.com/paper/2002.05540/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Feb 2020.

## Built on by (most cited first, 14 of 14)

- [Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning](https://mlsanity.com/paper/2101.10030/) (2021)
- [Skip-Convolutions for Efficient Video Processing](https://mlsanity.com/paper/2104.11487/) (2021)
- [SALISA: Saliency-based Input Sampling for Efficient Video Object Detection](https://mlsanity.com/paper/2204.02397/) (2022)
- [Gaining Scale Invariance in UAV Bird's Eye View Object Detection by Adaptive Resizing](https://mlsanity.com/paper/2101.12694/) (2021)
- [Incremental Multi-Target Domain Adaptation for Object Detection with Efficient Domain Transfer](https://mlsanity.com/paper/2104.06476/) (2021)
- [CenterPoly: real-time instance segmentation using bounding polygons](https://mlsanity.com/paper/2108.08923/) (2021)
- [FFAVOD: Feature Fusion Architecture for Video Object Detection](https://mlsanity.com/paper/2109.07298/) (2021)
- [A Comparison of Deep Saliency Map Generators on Multispectral Data in Object Detection](https://mlsanity.com/paper/2108.11767/) (2021)
- [Vehicle Detection and Tracking From Surveillance Cameras in Urban Scenes](https://mlsanity.com/paper/2109.12414/) (2021)
- [Scalable and Real-time Multi-Camera Vehicle Detection, Re-Identification, and Tracking](https://mlsanity.com/paper/2204.07442/) (2022)
- [Language-guided Learning for Object Detection Tackling Multiple Variations in Aerial Images](https://mlsanity.com/paper/2505.23193/) (2025)
- [Improving Token-based Object Detection with Video](https://mlsanity.com/paper/2506.22562/) (2025)
- [Localization-Based Tracking](https://mlsanity.com/paper/2104.05823/) (2021)
- [STF: Spatio-Temporal Fusion Module for Improving Video Object Detection](https://mlsanity.com/paper/2402.10752/) (2024)

## Abstract (arXiv)

Humans are very good at directing their visual attention toward relevant areas when they search for different types of objects. For instance, when we search for cars, we will look at the streets, not at the top of buildings. The motivation of this paper is to train a network to do the same via a multi-task learning approach. To train visual attention, we produce foreground/background segmentation labels in a semi-supervised way, using background subtraction or optical flow. Using these labels, we train an object detection model to produce foreground/background segmentation maps as well as bounding boxes while sharing most model parameters. We use those segmentation maps inside the network as a self-attention mechanism to weight the feature map used to produce the bounding boxes, decreasing the signal of non-relevant areas. We show that by using this method, we obtain a significant mAP improvement on two traffic surveillance datasets, with state-of-the-art results on both UA-DETRAC and UAVDT.