# Motion-Aware Reinforcement Learning For Object Localization

arXiv: https://arxiv.org/abs/2606.21764 · HTML: https://mlsanity.com/paper/2606.21764/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 19 of its 32 references that are in the map.
- First posted on arXiv 19 Jun 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 19 of 19)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [High-Dimensional Continuous Control Using Generalized Advantage Estimation](https://mlsanity.com/paper/1506.02438/) (2015)
- [Simple Online and Realtime Tracking](https://mlsanity.com/paper/1602.00763/) (2016)
- [Detection and Tracking Meet Drones Challenge](https://mlsanity.com/paper/2001.06303/) (2020)
- [Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Control](https://mlsanity.com/paper/1812.00568/) (2018)
- [Kinematic 3D Object Detection in Monocular Video](https://mlsanity.com/paper/2007.09548/) (2020)
- [Alpha-Refine: Boosting Tracking Performance by Precise Bounding Box Estimation](https://mlsanity.com/paper/2007.02024/) (2020)
- [Regularizing Action Policies for Smooth Control with Reinforcement Learning](https://mlsanity.com/paper/2012.06644/) (2020)
- [Active Object Localization with Deep Reinforcement Learning](https://mlsanity.com/paper/1511.06015/) (2015)
- [MotionTrack: Learning Motion Predictor for Multiple Object Tracking](https://mlsanity.com/paper/2306.02585/) (2023)
- [Gradient-based Regularization for Action Smoothness in Robotic Control with Reinforcement Learning](https://mlsanity.com/paper/2407.04315/) (2024)
- [Visible and Clear: Finding Tiny Objects in Difference Map](https://mlsanity.com/paper/2405.11276/) (2024)
- [Beyond Kalman Filters: Deep Learning-Based Filters for Improved Object Tracking](https://mlsanity.com/paper/2402.09865/) (2024)
- [Traffic Video Object Detection using Motion Prior](https://mlsanity.com/paper/2311.10092/) (2023)
- [Multi-Stage Reinforcement Learning For Object Detection](https://mlsanity.com/paper/1810.10325/) (2018)
- [Enhancing Your Trained DETRs with Box Refinement](https://mlsanity.com/paper/2307.11828/) (2023)

## Abstract (arXiv)

We present MARLNet (Motion-Aware Reinforcement Learning Network), a PPO-based bounding-box refinement agent that incorporates a constant-velocity motion prior into the observation state and an action smoothness penalty into the reward function. The agent operates on 268-dimensional observations encoding the current proposal, a kinematic prediction, the previous action, and a 256-dimensional EfficientNet-B0 crop feature, and learns a five-dimensional policy controlling coordinate adjustments and a binary termination trigger. Evaluated on Pascal VOC 2012 and VisDrone 2019, MARLNet trains stably across all regularization strengths tested and achieves consistent gains in detection success rate at IoU ≥ 0.5: up to +0.011 on VOC (λphys=0.10), where the motion prior prevents the overshooting that causes plain PPO to regress on this metric, and +0.007 on VisDrone (λphys=0.70), where unconstrained PPO achieves a larger gain (+0.025) owing to the weaker base detector. Through reward design ablations and training dynamics analysis, we identify a reward interference in which combining a constant-velocity deviation penalty with an absolute IoU term causes trigger collapse, and show that replacing it with the action smoothness penalty resolves this failure. We further characterize a representational ceiling facing crop-feature refinement agents that share a backbone with their base detector, confirmed through a global-plus-local observation ablation. Project page: https://prithviraj97.github.io/marl-net