Implementation:Datajuicer Data juicer VideoCameraCalibrationStaticDeepcalibMapper

Knowledge Sources	Datajuicer_Data_juicer
Domains	Data_Processing, Mapping
Last Updated	2026-02-14 16:00 GMT

Overview

Concrete tool for static camera calibration using DeepCalib provided by Data-Juicer.

Description

VideoCameraCalibrationStaticDeepcalibMapper is a mapper operator that computes camera intrinsics and field of view (FOV) for a static camera from video frames using the DeepCalib deep learning model. It extracts frames uniformly from videos, preprocesses them with Keras/ImageNet utilities, runs them through a DeepCalib regression model loaded from a .h5 weights file via TensorFlow/Keras, and stores the estimated camera calibration parameters in the sample metadata, with optional JSON output.

Usage

Use when you need automated camera calibration from video data to obtain intrinsic parameters for 3D reconstruction, undistortion, and other geometric video processing tasks.

Code Reference

Source Location

• Repository: Datajuicer_Data_juicer
• File: data_juicer/ops/mapper/video_camera_calibration_static_deepcalib_mapper.py

Signature

@OPERATORS.register_module("video_camera_calibration_static_deepcalib_mapper")
class VideoCameraCalibrationStaticDeepcalibMapper(Mapper):
    def __init__(self, model_path: str = "weights_10_0.02.h5", frame_num: PositiveInt = 3, duration: float = 0, tag_field_name: str = MetaKeys.static_camera_calibration_deepcalib_tags, frame_dir: str = DATA_JUICER_ASSETS_CACHE, if_output_info: bool = True, output_info_dir: str = DATA_JUICER_ASSETS_CACHE, *args, **kwargs):

Import

from data_juicer.ops.mapper.video_camera_calibration_static_deepcalib_mapper import VideoCameraCalibrationStaticDeepcalibMapper

I/O Contract

Inputs

Name	Type	Required	Description
model_path	str	No	Path to the DeepCalib regression model weights (default: "weights_10_0.02.h5")
frame_num	PositiveInt	No	Number of frames to extract uniformly (default: 3)
duration	float	No	Duration of each segment in seconds; 0 for entire video (default: 0)
tag_field_name	str	No	Field name to store tags (default: "static_camera_calibration_deepcalib_tags")
frame_dir	str	No	Output directory for extracted frames
if_output_info	bool	No	Whether to save camera parameters to JSON (default: True)
output_info_dir	str	No	Output directory for saving camera parameters JSON

Outputs

Name	Type	Description
samples	Dict	Transformed samples with camera intrinsics, distortion coefficients, and FOV metadata

Usage Examples

process:
  - video_camera_calibration_static_deepcalib_mapper:
      model_path: "weights_10_0.02.h5"
      frame_num: 5
      if_output_info: true

Related Pages

• Environment:Datajuicer_Data_juicer_Python_Runtime_Environment