This competition is so new to me on many levels. Predicting the behavior of other vehicles from bird's-eyes view is not what I anticipated from a self-driving system (I thought it would be more \"human\" - we don't usually imagine the surrounding from the 3rd view). Basically, this competition is about the 2nd stage on the schema below:
\nI hope that my work schedule will clear up so I can have some time to participate in this competition. Meanwhile, I will update this topic with my reading list for this challenge.
\nBehavior Prediction and Decision Making in Self-Driving Cars Using Deep Learning
\nA great Medium article that describes different approaches for motion prediction and decision making using Machine Learning and Deep Learning in self-driving cars. Nicely written and an easy read for beginners.
One Thousand and One Hours: Self-driving Motion Prediction Dataset [ArXiv:2006.14480]
\nThe ArXiv article that describes the dataset of this challenge (I think so). A must-read if you want to understand the data.
Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)
\nYou can watch this video while enjoying your lunch 😆😆😆
AI Papers are like veggies - you hate them but they're good for you.
\nUncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving
\nBy UBER. Published this year and presented on WACV'20. A deep learning-based approach that takes into account a current world state and produces raster images of each actor’s vicinity.
Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks
\nPublished on ICRA'19. A method to predict multiple possible trajectories of actors while also estimating their probabilities. The method encodes each actor’s surrounding context into a raster image, used as input by deep convolutional networks to automatically derive relevant features for the task.
I will keep this thread updated.
", "messages": [ { "id": "1014064", "postDate": "09/17/2020 07:21:55", "content": "This competition is so new to me on many levels. Predicting the behavior of other vehicles from bird's-eyes view is not what I anticipated from a self-driving system (I thought it would be more \"human\" - we don't usually imagine the surrounding from the 3rd view). Basically, this competition is about the 2nd stage on the schema below:
\nI hope that my work schedule will clear up so I can have some time to participate in this competition. Meanwhile, I will update this topic with my reading list for this challenge.
\nBehavior Prediction and Decision Making in Self-Driving Cars Using Deep Learning
\nA great Medium article that describes different approaches for motion prediction and decision making using Machine Learning and Deep Learning in self-driving cars. Nicely written and an easy read for beginners.
One Thousand and One Hours: Self-driving Motion Prediction Dataset [ArXiv:2006.14480]
\nThe ArXiv article that describes the dataset of this challenge (I think so). A must-read if you want to understand the data.
Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)
\nYou can watch this video while enjoying your lunch 😆😆😆
AI Papers are like veggies - you hate them but they're good for you.
\nUncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving
\nBy UBER. Published this year and presented on WACV'20. A deep learning-based approach that takes into account a current world state and produces raster images of each actor’s vicinity.
Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks
\nPublished on ICRA'19. A method to predict multiple possible trajectories of actors while also estimating their probabilities. The method encodes each actor’s surrounding context into a raster image, used as input by deep convolutional networks to automatically derive relevant features for the task.
I will keep this thread updated.
", "rawMarkdown": "This competition is so new to me on many levels. Predicting the behavior of other vehicles from bird's-eyes view is not what I anticipated from a self-driving system (I thought it would be more \"human\" - we don't usually imagine the surrounding from the 3rd view). Basically, this competition is about the 2nd stage on the schema below:\n\n\n\nI hope that my work schedule will clear up so I can have some time to participate in this competition. Meanwhile, I will update this topic with my reading list for this challenge.\n\n----------------------------------------------------------------\n\n# Introductory Articles - easy read for beginners\n\n- [**Behavior Prediction and Decision Making in Self-Driving Cars Using Deep Learning**](https://medium.com/@kargarisaac/behavior-prediction-and-decision-making-in-self-driving-cars-using-deep-learning-784761ed34af)\n A great Medium article that describes different approaches for motion prediction and decision making using Machine Learning and Deep Learning in self-driving cars. Nicely written and an easy read for beginners.\n\n- [**One Thousand and One Hours: Self-driving Motion Prediction Dataset [ArXiv:2006.14480]**](https://arxiv.org/abs/2006.14480)\n The ArXiv article that describes the dataset of this challenge (I think so). A must-read if you want to understand the data.\n\n- [**Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)**](https://www.youtube.com/watch?v=mxqdVO462HU)\n You can watch this video while enjoying your lunch 😆😆😆\n\n-------------------------------------------------------------------\n\n# Oh, the papers... Beginner's time is over!\n\nAI Papers are like veggies - you hate them but they're good for you.\n\n- [**Uncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving**](https://openaccess.thecvf.com/content_WACV_2020/papers/Djuric_Uncertainty-aware_Short-term_Motion_Prediction_of_Traffic_Actors_for_Autonomous_Driving_WACV_2020_paper.pdf)\n By UBER. Published this year and presented on WACV'20. A deep learning-based approach that takes into account a current world state and produces raster images of each actor’s vicinity.\n\n- [**Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks**](https://arxiv.org/abs/1809.10732?)\n Published on ICRA'19. A method to predict multiple possible trajectories of actors while also estimating their probabilities. The method encodes each actor’s surrounding context into a raster image, used as input by deep convolutional networks to automatically derive relevant features for the task.\n\n----------------------------------------------------------------\n\nI will keep this thread updated.", "votes": null }, { "id": "1018374", "postDate": "09/19/2020 16:36:36", "content": "Thanks a lot for sharing!
", "rawMarkdown": "Thanks a lot for sharing!", "votes": null } ], "comments": [ { "id": 1018374, "author_name": "amritvirsinghx", "author_url": "", "post_date": "09/19/2020 16:36:36", "content": "Thanks a lot for sharing!
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1014064": "This competition is so new to me on many levels. Predicting the behavior of other vehicles from bird's-eyes view is not what I anticipated from a self-driving system (I thought it would be more \"human\" - we don't usually imagine the surrounding from the 3rd view). Basically, this competition is about the 2nd stage on the schema below:\n\n\n\nI hope that my work schedule will clear up so I can have some time to participate in this competition. Meanwhile, I will update this topic with my reading list for this challenge.\n\n----------------------------------------------------------------\n\n# Introductory Articles - easy read for beginners\n\n- [**Behavior Prediction and Decision Making in Self-Driving Cars Using Deep Learning**](https://medium.com/@kargarisaac/behavior-prediction-and-decision-making-in-self-driving-cars-using-deep-learning-784761ed34af)\n A great Medium article that describes different approaches for motion prediction and decision making using Machine Learning and Deep Learning in self-driving cars. Nicely written and an easy read for beginners.\n\n- [**One Thousand and One Hours: Self-driving Motion Prediction Dataset [ArXiv:2006.14480]**](https://arxiv.org/abs/2006.14480)\n The ArXiv article that describes the dataset of this challenge (I think so). A must-read if you want to understand the data.\n\n- [**Teaching a Car to Drive Itself by Imitation and Imagination (Google I/O'19)**](https://www.youtube.com/watch?v=mxqdVO462HU)\n You can watch this video while enjoying your lunch 😆😆😆\n\n-------------------------------------------------------------------\n\n# Oh, the papers... Beginner's time is over!\n\nAI Papers are like veggies - you hate them but they're good for you.\n\n- [**Uncertainty-aware Short-term Motion Prediction of Traffic Actors for Autonomous Driving**](https://openaccess.thecvf.com/content_WACV_2020/papers/Djuric_Uncertainty-aware_Short-term_Motion_Prediction_of_Traffic_Actors_for_Autonomous_Driving_WACV_2020_paper.pdf)\n By UBER. Published this year and presented on WACV'20. A deep learning-based approach that takes into account a current world state and produces raster images of each actor’s vicinity.\n\n- [**Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks**](https://arxiv.org/abs/1809.10732?)\n Published on ICRA'19. A method to predict multiple possible trajectories of actors while also estimating their probabilities. The method encodes each actor’s surrounding context into a raster image, used as input by deep convolutional networks to automatically derive relevant features for the task.\n\n----------------------------------------------------------------\n\nI will keep this thread updated.", "1018374": "Thanks a lot for sharing!" }, "source": "meta" }