# Looking Fast and Slow: Memory-Guided Mobile Video Object Detection

arXiv: https://arxiv.org/abs/1903.10172 · HTML: https://mlsanity.com/paper/1903.10172/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2019.

## Built on by (most cited first, 22 of 22)

- [BEVDet4D: Exploit Temporal Cues in Multi-camera 3D Object Detection](https://mlsanity.com/paper/2203.17054/) (2022)
- [TransVOD: End-to-End Video Object Detection with Spatial-Temporal Transformers](https://mlsanity.com/paper/2201.05047/) (2022)
- [RegNet: Self-Regulated Network for Image Classification](https://mlsanity.com/paper/2101.00590/) (2021)
- [Tracking Pedestrian Heads in Dense Crowd](https://mlsanity.com/paper/2103.13516/) (2021)
- [Deep Learning for UAV-based Object Detection and Tracking: A Survey](https://mlsanity.com/paper/2110.12638/) (2021)
- [Skip-Convolutions for Efficient Video Processing](https://mlsanity.com/paper/2104.11487/) (2021)
- [COMISR: Compression-Informed Video Super-Resolution](https://mlsanity.com/paper/2105.01237/) (2021)
- [SALISA: Saliency-based Input Sampling for Efficient Video Object Detection](https://mlsanity.com/paper/2204.02397/) (2022)
- [Delta Distillation for Efficient Video Processing](https://mlsanity.com/paper/2203.09594/) (2022)
- [BlockCopy: High-Resolution Video Processing with Block-Sparse Feature Propagation and Online Policies](https://mlsanity.com/paper/2108.09376/) (2021)
- [Depth Estimation Matters Most: Improving Per-Object Depth Estimation for Monocular 3D Detection and Tracking](https://mlsanity.com/paper/2206.03666/) (2022)
- [PatchNet -- Short-range Template Matching for Efficient Video Processing](https://mlsanity.com/paper/2103.07371/) (2021)
- [Multi-resolution Rescored ByteTrack for Video Object Detection on Ultra-low-power Embedded Systems](https://mlsanity.com/paper/2404.11488/) (2024)
- [Real-Time Robust Video Object Detection System Against Physical-World Adversarial Attacks](https://mlsanity.com/paper/2208.09195/) (2022)
- [AI system for fetal ultrasound in low-resource settings](https://mlsanity.com/paper/2203.10139/) (2022)
- [BleedOrigin: Dynamic Bleeding Source Localization in Endoscopic Submucosal Dissection via Dual-Stage Detection and Tracking](https://mlsanity.com/paper/2507.15094/) (2025)
- [Memory Guided Road Detection](https://mlsanity.com/paper/2106.14184/) (2021)
- [Representation Recycling for Streaming Video Analysis](https://mlsanity.com/paper/2204.13492/) (2022)
- [SparseTem: Boosting the Efficiency of CNN-Based Video Encoders by Exploiting Temporal Continuity](https://mlsanity.com/paper/2410.20790/) (2024)
- [A Survey on Efficiency Optimization Techniques for DNN-based Video Analytics: Process Systems, Algorithms, and Applications](https://mlsanity.com/paper/2507.15628/) (2025)
- [MR2-ByteTrack: CNN and Transformer-based Video Object Detection for AI-augmented Embedded Vision Sensor Nodes](https://mlsanity.com/paper/2605.15423/) (2026)
- [Analytical and Convolutional Neural Network-Based Motion-Vector Propagation for Efficient Video Object Detection](https://mlsanity.com/paper/2609.34142/) (2026)

## Abstract (arXiv)

With a single eye fixation lasting a fraction of a second, the human visual system is capable of forming a rich representation of a complex environment, reaching a holistic understanding which facilitates object recognition and detection. This phenomenon is known as recognizing the "gist" of the scene and is accomplished by relying on relevant prior knowledge. This paper addresses the analogous question of whether using memory in computer vision systems can not only improve the accuracy of object detection in video streams, but also reduce the computation time. By interleaving conventional feature extractors with extremely lightweight ones which only need to recognize the gist of the scene, we show that minimal computation is required to produce accurate detections when temporal memory is present. In addition, we show that the memory contains enough information for deploying reinforcement learning algorithms to learn an adaptive inference policy. Our model achieves state-of-the-art performance among mobile methods on the Imagenet VID 2015 dataset, while running at speeds of up to 70+ FPS on a Pixel 3 phone.