In addition to the estimated disparity, the participants are required to submit a short technical report (maximum 4 pages, 10MB, pdf) summarizing their approach. The reports of all teams will be published on the website upon submission. The format of the report is left to the discretion of the participants, however, the report must specify the following information:
- A brief overview of the approach:
- Learning-based or optimization-based?
- How are events processed? E.g. event-by-event or using an event representation.
- Exact sensor modalities used (which event cameras and frame cameras)
- How much temporal information is incorporated? (window size of events, recurrency, multiple frames over time)
- Is the method causal? I.e. Does not use information from the future to predict at a given time.
- If a machine-learning-based approach is used:
- Was the model pre-trained? If yes, on which dataset(s)?
- Describe the data augmentation procedure, if it applies to your approach
The participants are welcome to include further details of their approach, eventual references to a paper describing the approach, or any other additional information.
Additional optional information could be:
- A detailed description of the approach, such as neural network architecture or optimization procedure.
- Loss function that was used for training/optimization.
- Expected processing time per prediction for a batch size of 1. In this case, also report the hardware (GPU/CPU/…) and software (Pytorch, TF, …) that you were using.
- Additional insights gained from working on this problem.