package waymo.open_dataset.occupancy_flow
Get desktop application:
View/edit binary Protocol Buffers messages
A set of ScenarioPrediction protos. A ScenarioPrediction proto for each example in the test or validation set must be included for a valid submission.
optional string account_name = 1
This must be set as the full email used to register at waymo.com/open.
optional string unique_method_name = 2
This name needs to be short, descriptive and unique. Only the latest result of the method from a user will show up on the leaderboard.
repeated string authors = 3
Author information.
optional string affiliation = 4
optional string description = 5
A brief description of the method.
optional string method_link = 6
Link to paper or other link that describes the method.
repeated scenario_predictions = 7
The set of scenario predictions to evaluate. One entry should exist for every record in the test set.
optional bool uses_lidar_data = 8
Set this to true if your model uses the lidar data provided in the motion dataset. This field is now REQUIRED for a valid submission.
optional bool uses_camera_data = 9
Set this to true if your model uses the camera data provided in the motion dataset. This field is now REQUIRED for a valid submission.
optional bool uses_public_model_pretraining = 10
Set this to true if your model used publicly available open-source LLM/VLM(s) for pre-training. This field is now REQUIRED for a valid submission.
repeated string public_model_names = 11
If any open-source model was used, specify their names and configuration.
optional string num_model_parameters = 12
Specify an estimate of the number of parameters of the model used to generate this submission. The number must be specified as an integer number followed by a multiplier suffix (from the set [K, M, B, T, ...], e.g. "200K"). This field is now REQUIRED for a valid submission.
A sequence of `num_waypoints` predictions for a single scenario.
Used in:
optional string scenario_id = 1
The unique ID of the scenario being predicted. This ID must match the scenario_id field in the test or validation set tf.Example or scenario proto corresponding to this set of predictions.
repeated waypoints = 2
Sequence of occupancy and flow predictions. This field should be repeated `num_waypoints` times.
Occupancy and flow predictions for a single waypoint (one-second interval).
Used in:
optional bytes observed_vehicles_occupancy = 1
Predicted row-major occupancy grid of currently-observed vehicles for one waypoint. It stores a numpy array shaped [grid_height, grid_width, 1] containing probabilites in range [0, 1] quantized to *unsigned int8* in range [0, 255]. To reduce the disk space requirements, the array is stored as zlib compressed bytes: Example code: predicted_occupancy = np.random.uniform(size=[256, 256, 1], low=0, high=1) predicted_occupancy = np.round(predicted_occupancy * 255).astype(np.uint8) bytes = zlib.compress(predicted_occupancy.tobytes())
optional bytes occluded_vehicles_occupancy = 2
Predicted occupancy grid of currently-occluded vehicles for one waypoint. Same representation as observed_vehicle_occupancy.
optional bytes all_vehicles_flow = 3
Predicted backward flow field of all currently-observed and currently- occluded vehicles for one waypoint. The flow vectors are in grid cell units and represent (dx, dy) displacements of points within agent boxes. Flow is constructed between timesteps `waypoint_size` apart. Please see occupancy_flow_grids.py for more detail. This field should store a numpy array shaped [grid_height, grid_width, 2] containing (dx, dy) vectors. The data is stored in row-major order. The values are quantized to *signed int8* in range [-128, 127]. To reduce the disk space requirements, the array is stored as zlib compressed bytes: Example code: predicted_flow = np.random.uniform(size=[256, 256, 2], low=-200, high=200) predicted_flow = np.clip(np.round(predicted_flow), -128, 127) predicted_flow = predicted_flow.astype(np.int8) bytes = zlib.compress(predicted_flow.tobytes())