{
  "fetched_at": "2026-10-20T00:24:02Z",
  "source_url": "https://www.kaggle.com/competitions/2nd-ava-challenge-ieee-mipr-2024/data",
  "pages": [
    {
      "name": "data-description",
      "content": "Your challenge in this competition is to forecast the car accident risk. In this dataset, we provide the videos of roads and freeways to let you train the model.\n\nPlease use training dataset to create your own model, and make classification on testing dataset. The name of each film is generated randomly. For your convenience, we named the road data after road_0000 to road_0299, while freeway is freeway_0000 to freeway_0301. Each film has a folder containing the corresponding jpg files.\n\n**5/24 update : The dataset has been updated. Please refer to the overview for detailed information.**\n\n## Files\n\n*   **2nd_AVA_DATASET_2ed**\n         |-------  **freeway**\n         |          |-------  **train** - the training videos of freeway\n         |          |-------  **test** - the testing videos of freeway\n         |-------  **road**\n         |          |-------  **train** - the training videos of road\n         |          |-------  **test** - the testing videos of road\n         |-------  **freeway_train.csv** - risk labels of videos in the freeway folder\n         |-------  **road_train.csv** - risk labels of videos in the road folder\n         |-------  **sample_submission.csv** - the example of submission format\n\n## Columns\n\n*   `film name` - name of videos (randomly generated)\n*   `risk` - the answer, represents the risk level of the car accident's impact on the recording vehicle. 1 indicates high risk, while 0 indicates low risk.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Description",
      "content": "Research in video analysis plays a vital role in autonomous driving by aiding in accident prediction. Dashcam footage provides valuable insights for this purpose. However, there's a lack of well-annotated, extensive datasets for accident prediction in Asian regions.\n\nTo address this issue, we collected a large-scale dataset of vehicle collisions from Taiwan and annotated it. The objective of this challenge is to predict the risk of an impending car accident to the recording vehicle.\n\n---\n#Final Award List\n**1st place** : Michael Ibrahim\n**2nd place** : MajurranVimalan-HumphreyMunn\n**3rd place** : ServerDown\n\nThe following groups have submitted their code and reports and have been verified without issues:\n| Rank | Team | Score  |\n| --- | --- | --- |\n| 1  |  Michael Ibrahim |0.7273 |\n| 2 |  MajurranVimalan-HumphreyMunn |0.7253 |\n| 6 |  ServerDown | 0.6576 |\n\n---\n\n# 7/7 Announcement: Important Information for the 2nd AVA Challenge\n\nThe deadline for submitting your technical report and code materials is approaching. Please ensure that the following tasks are completed:\n\n1. Send the links to your **documents(technical report and code materials)** via email to yunzhong1105@gmail.com and robert20000831@gmail.com .\n2. In your email, let us know if you are willing to attend the MIPR 2024 in-person presentation and award ceremony.\n\nPlease note: Competition rankings are provisional and subject to verification of the technical report before final confirmation. \n\nWe appreciate your efforts and look forward to seeing the winners at MIPR 2024!\n\n\n---\n\n# 7/1 Announcement: 2nd AVA Challenge Private Leaderboard Confirmed\n\nWe are pleased to announce the rankings for the 2nd AVA Challenge **Private Leaderboard**. The top six participants are as follows:\n\n| Rank | Team | Score  |\n| --- | --- | --- |\n| 1  |  Michael Ibrahim |0.7273 |\n| 2 |  MajurranVimalan-HumphreyMunn |0.7253 |\n| 3 |  TsaiTang00124 |0.7171 |\n| 4 |  Michele Comotti |0.6946 |\n| 5 |  Gaurav Dutta |0.6625 |\n| 5 |  Shubham76  |0.6625 |\n| 6 |  ServerDown | 0.6576 |\n\n \n\nCongratulations to these participants! To secure your rankings, please submit your technical report and code materials by 23:59 on July 7th(UTC+8). Detailed submission guidelines will be sent via email. Failure to submit the required documents by the deadline will result in disqualification, and the next participant in line will be promoted.\n\nThank you for your cooperation and participation. For more information, please contact us using the provided contact information.\n\n---\n\n# 5/24 Update\nDue to some labeling errors in the training dataset, we have uploaded the **2nd_AVA_Dataset_2ed** dataset with the following changes:\n\n* The number of frames for **freeway/train/freeway_0012** has increased from 80 to 136.\n\nPlease ensure that you have downloaded the correct version of the dataset.\n\n\n\n# Context\nAccording to the research, self-driving cars can only reduce about one-third of major car accidents that occur in the United States out of 5,000 incidents. People typically assume that autonomous driving systems can eliminate human errors while driving. However, many car accidents are still caused by human factors, including perception errors, driver distraction, poor visibility, or slow reactions. Other possible causes are prediction errors, such as misjudging the distance to other vehicles or pedestrians' movements. Some errors may be related to planning and decision-making, such as inappropriate evasive actions or overcompensation when controlling the vehicle. Other accidents can be caused by impairment due to alcohol, drugs, or drowsiness. There are avoidable accidents that are caused by vehicle malfunctions.\n\nBased on the above, we propose a topic for participants to utilize existing data to predict the risk of an impending car accident to the recording vehicle, in order to assist in crisis response for autonomous driving and enhance traffic safety.\n\nThe competition host is Advanced Computer Vision Lab (ACVLab), a research team led by Professor Chih-Chung Hsu at National Cheng Kung University. Our lab focuses on introducing machine learning/deep learning into various applications in computer vision and cultivating students with high practical and applied capabilities in deep learning.\n\nYour efforts in this competition could help extend the benefits of road safety. Greater access could further reduce the occurrence of car accidents worldwide.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Evaluation",
      "content": "## Goal\nThe goal of this competition is to use the given data to predict the risk of an accident involving a recorded vehicle. (Note: The risk variable represents the owner of the dashcam, not necessarily the vehicle that will have an accident.)\n\n## Metric\nThe evaluation metric for this competition is [**Roc Auc Score**](https://en.wikipedia.org/wiki/Receiver_operating_characteristic).\n\nThe ROC curve is the plot of the true positive rate (TPR) against the false positive rate (FPR) at each threshold setting. AUC means area under the curve of ROC.\n\nTPR is the proportion of samples that are correctly judged as positive among all samples that are actually positive. FPR means the proportion of samples that were incorrectly judged as positive among all samples that were actually negative. These two rate are given by:\n\n$$ \\text{TPR} = \\frac{\\text{TP}}{\\text{TP}+\\text{FN}}\\  ,\\ \\text{FPR} = \\frac{\\text{FP}}{\\text{FP}+\\text{TN}} $$\n\nA receiver operating characteristic curve, or ROC curve, is a graphical plot that illustrates the performance of a binary classifier model (which can also be used for multi-class classification) at varying threshold values. ROC analysis provides tools to select potentially optimal models and discard suboptimal ones independently from, and prior to specifying, the cost context or class distribution. It is directly and naturally related to the cost/benefit analysis of diagnostic decision making.\n\n## Submission Format\n\nPlease submit the file in .csv format that includes two features: `film name` and `risk`. You may refer to the dataset for a sample submission file.",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "rules",
      "content": "These are the rules of this competition. Participants must agree to the rules before joining. The rules are as follows:\n\n - **One account per participant** You cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.\n\n - **No private sharing outside teams** Privately sharing code or data outside of teams is not permitted. It's okay to share code if made available to all participants on the forums.\n\n - **Team Mergers** Team mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the Team Merger Deadline. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.\n\n - **Team Limits** The maximum team size is 5.\n\n - **Submission Limits** You may submit a maximum of 5 entries per day.\n\n - **You may select up to 2 final submissions for judging.**\n\n - **External Materials** The use of external materials that are publicly and freely available for research or academic purposes is permitted, but the source must be clearly stated in the technical report.\n\n - **Declaration about the Dataset** \n\n1. ACVLab makes no warranties regarding the Dataset, including but not limited to being up-to-date, correct or complete. ACVLab cannot be held liable for providing access to the Dataset or usage of the Dataset .\n2. The Dataset is only being used for scientific or research purposes. Any other use is explicitly prohibited. \n3. The Dataset must not be provided to any third party.\n4. The researcher takes full responsibility for usage of the Dataset and processing the Dataset.\n5. ACVLab reserves the right to terminate the researcher's access to the Dataset at any time.\n6. The place of jurisdiction is Taiwan.\n7. If any part of this agreement is legally invalid, this shall not affect the remaining agreement.\n \n - **Don't cheat!**",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Award & Recognition",
      "content": "After the competition concludes, we will invite the top 6 participants to submit technical reports/code for verification. Upon successful verification, we will then invite the three participants to co-work on this challenge summary paper, as well as to present (or virtually) their solutions at the IEEE MIPR2024 conference.\n\n**The technical report requires the MIPR paper format template, with a minimum of 2 pages.**\n\n \n\nOur sponsor, **PHISON**, is providing the following prizes :\n**1st place** : SSDs valued at approxiamtely $500\n**2nd place** : SSDs valued at approxiamtely $300\n**3rd place** : SSDs valued at approxiamtely $200",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Timeline",
      "content": "* **May 20, 2024** - Start Date\n\n* **June 30, 2024 11:55PM** - Final Submission Deadline\n\n* **July 1, 2024 12:00AM** - Private Leader Board Announce\n\n* **July 7, 2024** - Technical Report Submission, Completed Training Code and Related Materials Deadline\n\n* **July 14, 2024** - Competition End Date - Winner's announcement\n\nAll deadlines are at 11:59 PM UTC+8 on a corresponding day unless otherwise noted. The competition organizers reserve the right to update the contest timeline if they deem it necessary.\n",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Contact Information",
      "content": "If you have any questions, please contact the following ACVLab members:\nYun-Zhong Jiang (yunzhong1105@gmail.com)\nWei-Hao Huang (robert20000831@gmail.com)\nYu-Fan Lin (aas12as12as12tw@gmail.com)",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    },
    {
      "name": "Organizers",
      "content": "**Organization Chair**\n* Prof. Chih-Chung Hsu - Institute of Data Science, National Cheng Kung University, Tainan, Taiwan\n\n* Prof. Ming-Ching Chang - Department of Computer Science, University at Albany, State University of New York, USA\n\n**Organization Committee**\n- Yun-Zhong Jiang (yunzhong1105@gmail.com)\n- Wei-Hao Huang (robert20000831@gmail.com)\n- Yu-Fan Ling (aas12as12as12tw@gmail.com)",
      "isPublished": false,
      "hasIsPublished": false,
      "postTitle": "",
      "hasPostTitle": false,
      "mimeType": "",
      "hasMimeType": false
    }
  ]
}