Improving Generalization of Transfer Learning Across Domains Using Spatio-Temporal Features in Autonomous Driving

• URL: http://arxiv.org/abs/2103.08116v1
• Date: Mon, 15 Mar 2021 03:26:06 GMT
• Title: Improving Generalization of Transfer Learning Across Domains Using Spatio-Temporal Features in Autonomous Driving
• Authors: Shivam Akhauri, Laura Zheng, Tom Goldstein, Ming Lin
• Abstract summary: Vehicle simulation can be used to learn in the virtual world, and the acquired skills can be transferred to handle real-world scenarios. These visual elements are intuitively crucial for human decision making during driving. We propose a CNN+LSTM transfer learning framework to extract thetemporal-temporal features representing vehicle dynamics from scenes.
• Score: 45.655433907239804
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training vision-based autonomous driving in the real world can be inefficient and impractical. Vehicle simulation can be used to learn in the virtual world, and the acquired skills can be transferred to handle real-world scenarios more effectively. Between virtual and real visual domains, common features such as relative distance to road edges and other vehicles over time are consistent. These visual elements are intuitively crucial for human decision making during driving. We hypothesize that these spatio-temporal factors can also be used in transfer learning to improve generalization across domains. First, we propose a CNN+LSTM transfer learning framework to extract the spatio-temporal features representing vehicle dynamics from scenes. Next, we conduct an ablation study to quantitatively estimate the significance of various features in the decisions of driving systems. We observe that physically interpretable factors are highly correlated with network decisions, while representational differences between scenes are not. Finally, based on the results of our ablation study, we propose a transfer learning pipeline that uses saliency maps and physical features extracted from a source model to enhance the performance of a target model. Training of our network is initialized with the learned weights from CNN and LSTM latent features (capturing the intrinsic physics of the moving vehicle w.r.t. its surroundings) transferred from one domain to another. Our experiments show that this proposed transfer learning framework better generalizes across unseen domains compared to a baseline CNN model on a binary classification learning task.

Related papers

• Transfer Learning Study of Motion Transformer-based Trajectory Predictions [1.2972104025246092]
  Trajectory planning in autonomous driving is highly dependent on predicting the emergent behavior of other road users. Learning-based methods are currently showing impressive results in simulation-based challenges. The study aims to provide insights into possible trade-offs between computational time and performance to support effective transfers into the real world.
  arXiv  Detail & Related papers  (2024-04-12T06:50:32Z)
• KARNet: Kalman Filter Augmented Recurrent Neural Network for Learning World Models in Autonomous Driving Tasks [11.489187712465325]
  We present a Kalman filter augmented recurrent neural network architecture to learn the latent representation of the traffic flow using front camera images only. Results show that incorporating an explicit model of the vehicle (states estimated using Kalman filtering) in the end-to-end learning significantly increases performance.
  arXiv  Detail & Related papers  (2023-05-24T02:27:34Z)
• TrafficBots: Towards World Models for Autonomous Driving Simulation and Motion Prediction [149.5716746789134]
  We show data-driven traffic simulation can be formulated as a world model. We present TrafficBots, a multi-agent policy built upon motion prediction and end-to-end driving. Experiments on the open motion dataset show TrafficBots can simulate realistic multi-agent behaviors.
  arXiv  Detail & Related papers  (2023-03-07T18:28:41Z)
• CARNet: A Dynamic Autoencoder for Learning Latent Dynamics in Autonomous Driving Tasks [11.489187712465325]
  An autonomous driving system should effectively use the information collected from the various sensors in order to form an abstract description of the world. Deep learning models, such as autoencoders, can be used for that purpose, as they can learn compact latent representations from a stream of incoming data. This work proposes CARNet, a Combined dynAmic autoencodeR NETwork architecture that utilizes an autoencoder combined with a recurrent neural network to learn the current latent representation.
  arXiv  Detail & Related papers  (2022-05-18T04:15:42Z)
• Towards Scale Consistent Monocular Visual Odometry by Learning from the Virtual World [83.36195426897768]
  We propose VRVO, a novel framework for retrieving the absolute scale from virtual data. We first train a scale-aware disparity network using both monocular real images and stereo virtual data. The resulting scale-consistent disparities are then integrated with a direct VO system.
  arXiv  Detail & Related papers  (2022-03-11T01:51:54Z)
• Towards Optimal Strategies for Training Self-Driving Perception Models in Simulation [98.51313127382937]
  We focus on the use of labels in the synthetic domain alone. Our approach introduces both a way to learn neural-invariant representations and a theoretically inspired view on how to sample the data from the simulator. We showcase our approach on the bird's-eye-view vehicle segmentation task with multi-sensor data.
  arXiv  Detail & Related papers  (2021-11-15T18:37:43Z)
• TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviors [74.67698916175614]
  We propose TrafficSim, a multi-agent behavior model for realistic traffic simulation. In particular, we leverage an implicit latent variable model to parameterize a joint actor policy. We show TrafficSim generates significantly more realistic and diverse traffic scenarios as compared to a diverse set of baselines.
  arXiv  Detail & Related papers  (2021-01-17T00:29:30Z)
• Explicit Domain Adaptation with Loosely Coupled Samples [85.9511585604837]
  We propose a transfer learning framework, core of which is learning an explicit mapping between domains. Due to its interpretability, this is beneficial for safety-critical applications, like autonomous driving.
  arXiv  Detail & Related papers  (2020-04-24T21:23:45Z)

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.