video¶

psifx video¶

Command-line interface for processing videos.

usage: psifx video [-h] [--all-help] {manipulation,pose,face,tracking} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video face¶

Command-line interface for estimating face features from videos.

usage: psifx video face [-h] [--all-help] {openface} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video face openface¶

Command-line interface for running OpenFace.

usage: psifx video face openface [-h] [--all-help]
                                 {single-inference,multi-inference,visualization}
                                 ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video face openface multi-inference¶

Command-line interface for inferring face features from videos with OpenFace.

usage: psifx video face openface multi-inference [-h] --video VIDEO --masks
                                                 MASKS [MASKS ...]
                                                 --features_dir FEATURES_DIR
                                                 [--device DEVICE]
                                                 [--overwrite | --no-overwrite]
                                                 [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--masks <masks>¶: list of path to mask directories or individual .mp4 mask files

--features_dir <features_dir>¶: path to the output feature directory, such as /path/to/openface

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video face openface single-inference¶

Command-line interface for inferring face features from videos with OpenFace.

usage: psifx video face openface single-inference [-h] --video VIDEO
                                                  --features FEATURES
                                                  [--mask MASK]
                                                  [--device DEVICE]
                                                  [--overwrite | --no-overwrite]
                                                  [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--features <features>¶: path to the output feature archive, such as /path/to/openface.tar.gz

--mask <mask>¶: path to the input .mp4 mask file

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video face openface visualization¶

Command-line interface for visualizing face features from videos with OpenFace.

usage: psifx video face openface visualization [-h] --video VIDEO --features
                                               FEATURES [FEATURES ...]
                                               --visualization VISUALIZATION
                                               [--depth DEPTH] [--f_x F_X]
                                               [--f_y F_Y] [--c_x C_X]
                                               [--c_y C_Y]
                                               [--overwrite | --no-overwrite]
                                               [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--features <features>¶: list of path to the input feature directories or individual archive .tar.gz files

--visualization <visualization>¶: path to the output video file, such as /path/to/visualization.mp4 (or .avi, .mkv, etc.)

--depth <depth>¶: projection: assumed static depth of the subject in meters

--f_x <f_x>¶: projection: x-axis of the focal length

--f_y <f_y>¶: projection: y-axis of the focal length

--c_x <c_x>¶: projection: x-axis of the principal point

--c_y <c_y>¶: projection: y-axis of the principal point

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video manipulation¶

Command-line interface for manipulating videos.

usage: psifx video manipulation [-h] [--all-help] {process} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video manipulation process¶

Command-line interface for processing videos. The trimming, cropping and resizing can be performed all at once, and in that order.

usage: psifx video manipulation process [-h] --in_video IN_VIDEO --out_video
                                        OUT_VIDEO [--start START] [--end END]
                                        [--x_min X_MIN] [--y_min Y_MIN]
                                        [--x_max X_MAX] [--y_max Y_MAX]
                                        [--width WIDTH] [--height HEIGHT]
                                        [--overwrite | --no-overwrite]
                                        [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--in_video <in_video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--out_video <out_video>¶: path to the output video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--start <start>¶: trim: timestamp in seconds of the start of the selection

--end <end>¶: trim: timestamp in seconds of the end of the selection

--x_min <x_min>¶: crop: x-axis coordinate of the top-left corner in pixels

--y_min <y_min>¶: crop: y-axis coordinate of the top-left corner in pixels

--x_max <x_max>¶: crop: x-axis coordinate of the bottom-right corner in pixels

--y_max <y_max>¶: crop: y-axis coordinate of the bottom-right corner in pixels

--width <width>¶: resize: width of the resized output

--height <height>¶: resize: height of the resized output

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video pose¶

Command-line interface for estimating human poses from videos.

usage: psifx video pose [-h] [--all-help] {mediapipe,visualization} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video pose mediapipe¶

Command-line interface for running MediaPipe.

usage: psifx video pose mediapipe [-h] [--all-help]
                                  {single-inference,multi-inference,visualization}
                                  ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video pose mediapipe multi-inference¶

Command-line interface for inferring human pose with MediaPipe Holistic.

usage: psifx video pose mediapipe multi-inference [-h] --video VIDEO
                                                  --poses_dir POSES_DIR
                                                  --masks MASKS [MASKS ...]
                                                  [--mask_threshold MASK_THRESHOLD]
                                                  [--model_complexity MODEL_COMPLEXITY]
                                                  [--smooth | --no-smooth]
                                                  [--device DEVICE]
                                                  [--overwrite | --no-overwrite]
                                                  [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--poses_dir <poses_dir>¶: path to the output pose directory, such as /path/to/poses

--masks <masks>¶: list of path to mask directories or individual .mp4 mask files

--mask_threshold <mask_threshold>¶: threshold for the binarization of the segmentation mask

--model_complexity <model_complexity>¶: complexity of the model: {0, 1, 2}, higher means more FLOPs, but also more accurate results

--smooth, --no-smooth¶: temporally smooth the inference results to reduce the jitter

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video pose mediapipe single-inference¶

Command-line interface for inferring human pose with MediaPipe Holistic.

usage: psifx video pose mediapipe single-inference [-h] --video VIDEO --poses
                                                   POSES [--mask MASK]
                                                   [--mask_threshold MASK_THRESHOLD]
                                                   [--model_complexity MODEL_COMPLEXITY]
                                                   [--smooth | --no-smooth]
                                                   [--device DEVICE]
                                                   [--overwrite | --no-overwrite]
                                                   [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--poses <poses>¶: path to the output pose archive, such as /path/to/poses.tar.gz

--mask <mask>¶: path to the input .mp4 mask file

--mask_threshold <mask_threshold>¶: threshold for the binarization of the segmentation mask

--model_complexity <model_complexity>¶: complexity of the model: {0, 1, 2}, higher means more FLOPs, but also more accurate results

--smooth, --no-smooth¶: temporally smooth the inference results to reduce the jitter

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video pose mediapipe visualization¶

Command-line interface for visualizing the poses over the video.

usage: psifx video pose mediapipe visualization [-h] --video VIDEO --poses
                                                POSES [POSES ...]
                                                --visualization VISUALIZATION
                                                [--confidence_threshold CONFIDENCE_THRESHOLD]
                                                [--overwrite | --no-overwrite]
                                                [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--poses <poses>¶: list of path to the input pose directories or individual archive .tar.gz files

--visualization <visualization>¶: path to the output visualization video file, such as /path/to/visualization.mp4 (or .avi, .mkv, etc.)

--confidence_threshold <confidence_threshold>¶: threshold for not displaying low confidence keypoints

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video pose visualization¶

Command-line interface for visualizing the poses over the video.

usage: psifx video pose visualization [-h] --video VIDEO --poses POSES
                                      [POSES ...] --visualization
                                      VISUALIZATION
                                      [--confidence_threshold CONFIDENCE_THRESHOLD]
                                      [--overwrite | --no-overwrite]
                                      [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--poses <poses>¶: list of path to the input pose directories or individual archive .tar.gz files

--visualization <visualization>¶: path to the output visualization video file, such as /path/to/visualization.mp4 (or .avi, .mkv, etc.)

--confidence_threshold <confidence_threshold>¶: threshold for not displaying low confidence keypoints

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video tracking¶

Command-line interface for processing videos.

usage: psifx video tracking [-h] [--all-help] {sam3,visualization} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video tracking sam3¶

Command-line interface for running SAM3.

usage: psifx video tracking sam3 [-h] [--all-help]
                                 {inference,visualization} ...

-h, --help¶: show this help message and exit

--all-help¶: show help recursively and exit

psifx video tracking sam3 inference¶

Command-line interface for tracking video elements with SAM3.

usage: psifx video tracking sam3 inference [-h] --video VIDEO --mask_dir
                                           MASK_DIR
                                           [--text_prompt TEXT_PROMPT]
                                           [--chunk_size CHUNK_SIZE]
                                           [--iou_threshold IOU_THRESHOLD]
                                           [--max_num_objects MAX_NUM_OBJECTS]
                                           [--device DEVICE]
                                           [--model_path MODEL_PATH]
                                           [--api_token API_TOKEN]
                                           [--overwrite | --no-overwrite]
                                           [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--mask_dir <mask_dir>¶: path to the output mask directory, such as /path/to/mask_dir

--text_prompt <text_prompt>¶: text description of objects to track (e.g., ‘people’, ‘cars’, ‘dogs’)

--chunk_size <chunk_size>¶: number of frames to process at once (lower values use less memory)

--iou_threshold <iou_threshold>¶: IoU threshold for stitching chunks together (0.0 to 1.0)

--max_num_objects <max_num_objects>¶: optional cap on tracked object count (e.g., set to 2 for a two-person interaction)

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--model_path <model_path>¶: SAM3 model id or local path (e.g. ‘facebook/sam3’ or ‘/path/to/sam3’)

--api_token <api_token>¶: Hugging Face token (defaults to HF_TOKEN env var when available)

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video tracking sam3 visualization¶

Command-line interface for the visualization of tracking.

usage: psifx video tracking sam3 visualization [-h] --video VIDEO --masks
                                               MASKS [MASKS ...]
                                               --visualization VISUALIZATION
                                               [--blackout | --no-blackout]
                                               [--labels | --no-labels]
                                               [--color | --no-color]
                                               [--device DEVICE]
                                               [--overwrite | --no-overwrite]
                                               [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--masks <masks>¶: list of path to mask directories or individual .mp4 mask files

--visualization <visualization>¶: path to the output visualization video file, such as /path/to/visualization.mp4 (or .avi, .mkv, etc.)

--blackout, --no-blackout¶: whether to black out the background (non-mask regions)

--labels, --no-labels¶: whether to add labels

--color, --no-color¶: whether to color the masks

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script

psifx video tracking visualization¶

Command-line interface for the visualization of tracking.

usage: psifx video tracking visualization [-h] --video VIDEO --masks MASKS
                                          [MASKS ...] --visualization
                                          VISUALIZATION
                                          [--blackout | --no-blackout]
                                          [--labels | --no-labels]
                                          [--color | --no-color]
                                          [--device DEVICE]
                                          [--overwrite | --no-overwrite]
                                          [--verbose | --no-verbose]

-h, --help¶: show this help message and exit

--video <video>¶: path to the input video file, such as /path/to/video.mp4 (or .avi, .mkv, etc.)

--masks <masks>¶: list of path to mask directories or individual .mp4 mask files

--visualization <visualization>¶: path to the output visualization video file, such as /path/to/visualization.mp4 (or .avi, .mkv, etc.)

--blackout, --no-blackout¶: whether to black out the background (non-mask regions)

--labels, --no-labels¶: whether to add labels

--color, --no-color¶: whether to color the masks

--device <device>¶: device on which to run the inference, either ‘cpu’ or ‘cuda’

--overwrite, --no-overwrite¶: overwrite existing files, otherwise raises an error

--verbose, --no-verbose¶: verbosity of the script