# Lightweight Multi-Frame Integration for Robust YOLO Object Detection in Videos

arXiv: https://arxiv.org/abs/2506.20550 · HTML: https://mlsanity.com/paper/2506.20550/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 27 references that are in the map.
- First posted on arXiv 25 Jun 2025.

## Builds on (most cited first, 18 of 18)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [You Only Look Once: Unified, Real-Time Object Detection](https://mlsanity.com/paper/1506.02640/) (2015)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors](https://mlsanity.com/paper/2207.02696/) (2022)
- [Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/1710.11063/) (2017)
- [MOT20: A benchmark for multi object tracking in crowded scenes](https://mlsanity.com/paper/2003.09003/) (2020)
- [Eigen-CAM: Class Activation Map using Principal Components](https://mlsanity.com/paper/2008.00299/) (2020)
- [Flow-Guided Feature Aggregation for Video Object Detection](https://mlsanity.com/paper/1703.10025/) (2017)
- [Sequence Level Semantics Aggregation for Video Object Detection](https://mlsanity.com/paper/1907.06390/) (2019)
- [SeaDronesSee: A Maritime Benchmark for Detecting Humans in Open Water](https://mlsanity.com/paper/2105.01922/) (2021)
- [Video Object Detection with an Aligned Spatial-Temporal Memory](https://mlsanity.com/paper/1712.06317/) (2017)
- [YOLOV: Making Still Image Object Detectors Great at Video Object Detection](https://mlsanity.com/paper/2208.09686/) (2022)
- [Label-Efficient Online Continual Object Detection in Streaming Video](https://mlsanity.com/paper/2206.00309/) (2022)
- [1st Workshop on Maritime Computer Vision (MaCVi) 2023: Challenge Results](https://mlsanity.com/paper/2211.13508/) (2022)
- [An Efficient 3D CNN for Action/Object Segmentation in Video](https://mlsanity.com/paper/1907.08895/) (2019)
- [Recurrent Neural Networks for video object detection](https://mlsanity.com/paper/2010.15740/) (2020)

## Built on by (most cited first, 4 of 4)

- [A Review of Vision-Based Vehicle Detection for UAV-Based Traffic Monitoring: Experimental Insights and Future Directions](https://mlsanity.com/paper/2608.07571/) (2026)
- [Detection and Identification of Penguins Using Appearance and Motion Features](https://mlsanity.com/paper/2603.03603/) (2026)
- [Video-based Locomotion Analysis for Fish Health Monitoring](https://mlsanity.com/paper/2603.05407/) (2026)
- [Non-invasive Growth Monitoring of Small Freshwater Fish in Home Aquariums via Stereo Vision](https://mlsanity.com/paper/2603.06421/) (2026)

## Abstract (arXiv)

Modern image-based object detection models, such as YOLOv7, primarily process individual frames independently, thus ignoring valuable temporal context naturally present in videos. Meanwhile, existing video-based detection methods often introduce complex temporal modules, significantly increasing model size and computational complexity. In practical applications such as surveillance and autonomous driving, transient challenges including motion blur, occlusions, and abrupt appearance changes can severely degrade single-frame detection performance. To address these issues, we propose a straightforward yet highly effective strategy: stacking multiple consecutive frames as input to a YOLO-based detector while supervising only the output corresponding to a single target frame. This approach leverages temporal information with minimal modifications to existing architectures, preserving simplicity, computational efficiency, and real-time inference capability. Extensive experiments on the challenging MOT20Det and our BOAT360 datasets demonstrate that our method improves detection robustness, especially for lightweight models, effectively narrowing the gap between compact and heavy detection networks. Additionally, we contribute the BOAT360 benchmark dataset, comprising annotated fisheye video sequences captured from a boat, to support future research in multi-frame video object detection in challenging real-world scenarios.