MCPcopy Create free account

hub / github.com/Netflix/void-model / functions

Functions673 in github.com/Netflix/void-model

↓ 1 callersMethodcheck_inputs
( self, prompt, height, width, negative_prompt, callback_on_st
videox_fun/pipeline/pipeline_wan_fun.py:314
↓ 1 callersMethodcheck_inputs
( self, prompt, height, width, negative_prompt, callback_on_st
videox_fun/pipeline/pipeline_cogvideox_fun_inpaint.py:604
↓ 1 callersFunctionclear_scene
Remove all objects from the scene.
data_generation/render_paired_videos_blender_quadmask.py:162
↓ 1 callersFunctioncolor_transfer
Transfer color distribution from of sc, referred to dc. Args: sc (numpy.ndarray): input image to be transfered. dc (numpy.nd
videox_fun/utils/utils.py:37
↓ 1 callersFunctioncombine_masks
Combine black and grey mask frames. Rules: - black=0, grey=255 → 0 (primary object only) - black=255, grey=127 → 127 (affected objec
VLM-MASK-REASONER/stage4_combine_masks.py:23
↓ 1 callersMethodcompute_joint_positions
Compute the positions of all joints in world space Args: pose_params: Dict of bone_name -> tensor [batch_size, 4, 4] (rot
data_generation/human_model/human_model.py:165
↓ 1 callersFunctionconvert_mask_to_hybrid_grid
Convert quad mask to hybrid grid-aligned mask. Process: 1. Grid-align 127 (grey): cell gets 127 if ANY pixel is 127, else 255 2. Kee
data_generation/convert_masks_to_grid_hybrid.py:61
↓ 1 callersFunctionconvert_model_weight_to_float8
(model, exclude_module_name=['embed_tokens'])
videox_fun/utils/fp8_optimization.py:28
↓ 1 callersFunctionconvert_sequence_for_character
Convert one sequence for a specific character. Runs the 4-step conversion process.
data_generation/convert_all_scenarios_for_characters.py:72
↓ 1 callersFunctionconvert_v2_5_from_siglip
( predictor_name_or_path: str | PathLike | None = None, encoder_model_name: str = "google/siglip-so400
videox_fun/reward/aesthetic_predictor_v2_5/siglip_v2_5.py:109
↓ 1 callersFunctioncreate_first_frame_with_mask_overlay
Create visualization of first frame with red overlay on primary object Args: first_frame_path: Path to first_frame.jpg black_mask
VLM-MASK-REASONER/stage2_vlm_analysis_cf.py:71
↓ 1 callersFunctioncreate_first_frame_with_mask_overlay
Create visualization of first frame with red overlay on primary object Args: first_frame_path: Path to first_frame.jpg black_mask
VLM-MASK-REASONER/stage2_vlm_analysis.py:70
↓ 1 callersFunctioncreate_ground_plane
Create a ground plane with optional PBR textures or chessboard pattern. Args: size: Size of the plane y_up: Whether using Y-up co
data_generation/render_paired_videos_blender_quadmask.py:249
↓ 1 callersFunctioncreate_human_mesh
Create human mesh from pre-computed vertices and faces. Args: vertices: numpy array [num_verts, 3] - first frame vertices fa
data_generation/render_paired_videos_blender_quadmask.py:653
↓ 1 callersMethodcreate_mask_visualization
Create RGB visualization of mask with color coding and RGB background
VLM-MASK-REASONER/edit_quadmask.py:341
↓ 1 callersFunctioncreate_multi_frame_grid_samples
Create gridded frame samples at multiple time points in video. Helps VLM see objects that appear mid-video with grid reference. Args:
VLM-MASK-REASONER/stage2_vlm_analysis_cf.py:180
↓ 1 callersFunctioncreate_multi_frame_grid_samples
Create gridded frame samples at multiple time points in video. Helps VLM see objects that appear mid-video with grid reference. Args:
VLM-MASK-REASONER/stage2_vlm_analysis.py:179
↓ 1 callersFunctioncreate_quadmask
Create quad-mask from tri-mask and RGB frames. Args: mask_frame: Original mask frame (numpy array) rgb_full_frame: Full RGB
data_generation/convert_trimask_to_quadmask.py:36
↓ 1 callersFunctioncreate_trimask_from_frames
Build quad-mask using 4 renders: 0 (black) : human only (no scene change underneath) 63 (dark gray) : overlap (human coverin
data_generation/render_paired_videos_blender_quadmask.py:1022
↓ 1 callersFunctioncreate_video_from_frames
Create video from frames using ffmpeg.
data_generation/convert_trimask_to_quadmask.py:102
↓ 1 callersFunctioncreate_video_from_frames
Create grayscale video from frames using ffmpeg.
data_generation/convert_masks_to_grid_hybrid.py:140
↓ 1 callersFunctioncreate_walls
Create 4 walls around the scene with optional PBR textures. Args: size: Size of the room (walls will be at +/- size/2) height: He
data_generation/render_paired_videos_blender_quadmask.py:473
↓ 1 callersFunctiondeepspeed_zero_init_disabled_context_manager
returns either a context list that includes one that will disable zero.Init or an empty context list
scripts/cogvideox_fun/train.py:808
↓ 1 callersFunctiondeepspeed_zero_init_disabled_context_manager
returns either a context list that includes one that will disable zero.Init or an empty context list
scripts/cogvideox_fun/train_warped_noise.py:826
↓ 1 callersFunctiondefault
(val, d)
videox_fun/reward/MPS/trainer/models/cross_modeling.py:11
↓ 1 callersFunctiondilate_mask
Dilate mask for proximity checking
VLM-MASK-REASONER/stage3a_generate_grey_masks_v2.py:319
↓ 1 callersFunctiondilate_mask
Dilate mask to create proximity region
VLM-MASK-REASONER/stage3a_generate_grey_masks.py:178
↓ 1 callersMethoddraw_grid_overlay
Draw grid lines on mask canvas
VLM-MASK-REASONER/edit_quadmask.py:401
↓ 1 callersMethodencode_prompt
r""" Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*):
videox_fun/pipeline/pipeline_cogvideox_fun.py:331
↓ 1 callersMethodencode_prompt
r""" Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*):
videox_fun/pipeline/pipeline_wan_fun.py:180
↓ 1 callersMethodencode_prompt
r""" Encodes the prompt into text encoder hidden states. Args: prompt (`str` or `List[str]`, *optional*):
videox_fun/pipeline/pipeline_cogvideox_fun_inpaint.py:389
↓ 1 callersMethodextract_flow
Extract optical flow between two consecutive frames. Args: frame1: First frame [B, C, H, W] in range [0, 1] or [0, 255]
videox_fun/utils/optical_flow_utils.py:74
↓ 1 callersFunctionextract_frames
Extract all frames from a video using ffmpeg.
data_generation/convert_masks_to_grid_hybrid.py:124
↓ 1 callersFunctionextract_hip_position
Extract hip positions from armature data for camera placement. Args: armature_data: List of frame bone dictionaries bone_nam
data_generation/blender_utils.py:302
↓ 1 callersFunctionextract_object_data
()
data_generation/scripts/extract_pk_data.py:59
↓ 1 callersFunctionextract_object_model_data
Extract mesh vertices and faces from objects. Works in both object mode and edit mode.
data_generation/scripts/extract_pk_data.py:79
↓ 1 callersFunctionextract_skeleton_data
()
data_generation/scripts/extract_pk_data.py:31
↓ 1 callersMethodfake_context_parallel_forward
( self, inputs: torch.Tensor, conv_cache: Optional[torch.Tensor] = None )
videox_fun/models/cogvideox_vae.py:124
↓ 1 callersFunctionffmpeg_exe
()
data_generation/kubric_variable_objects.py:132
↓ 1 callersFunctionfill_mask
(shape, x, y, fill_val=255)
videox_fun/data/dataset_image_video.py:114
↓ 1 callersFunctionfill_mask
(shape, x, y, fill_val=255)
videox_fun/data/dataset_image_video_warped.py:114
↓ 1 callersFunctionfilter_by_proximity
Filter mask to only include regions near primary mask
VLM-MASK-REASONER/stage3a_generate_grey_masks.py:184
↓ 1 callersFunctionfilter_kwargs
(cls, kwargs)
videox_fun/utils/utils.py:23
↓ 1 callersFunctionfilter_masks_by_proximity
Filter masks to only include regions near primary mask
VLM-MASK-REASONER/stage3a_generate_grey_masks_v2.py:325
↓ 1 callersFunctionfind_objects_needing_trajectory
Find all objects that need trajectory input
VLM-MASK-REASONER/stage3b_trajectory_gui.py:333
↓ 1 callersFunctionfind_pass1_video
Find the pass 1 video for a given video name. Args: video_name: Base name (e.g., "ball-play") pass1_dir: Directory containin
inference/cogvideox_fun/inference_with_pass1_warped_noise.py:48
↓ 1 callersFunctionfind_sequences_with_verts
Find all sequences with pre-computed vertices.
data_generation/generate_review_frames.py:18
↓ 1 callersFunctionfind_sequences_with_verts
Find all sequences with pre-computed vertices.
data_generation/generate_review_frames_direct.py:28
↓ 1 callersFunctionfind_video_folders
Find all folders containing the required video files.
data_generation/convert_trimask_to_quadmask.py:224
↓ 1 callersFunctionfind_video_folders
Find all folders containing mask videos.
data_generation/convert_masks_to_grid_hybrid.py:284
↓ 1 callersMethodfinish
All objects done
VLM-MASK-REASONER/stage3b_trajectory_gui.py:326
↓ 1 callersFunctionflow_to_motion_rgb
Stable flow->RGB with fixed pixel clip for consistent visualization. NOTE: This is for VISUALIZATION, not VideoJAM training! For VideoJA
videox_fun/utils/optical_flow_utils.py:337
↓ 1 callersFunctionflow_to_motion_rgb_videojam
Convert optical flow to RGB motion representation following VideoJAM paper specification. This is NOT a visualization - it's a weak, normali
videox_fun/utils/optical_flow_utils.py:211
↓ 1 callersMethodforward
(self, image_embeds: torch.Tensor)
videox_fun/reward/aesthetic_predictor_v2_5/siglip_v2_5.py:41
↓ 1 callersFunctiongenerate_one
(rng, out_dir: Path, kubasic_source, gso_source, hdri_source, objects_split=
data_generation/kubric_variable_objects.py:162
↓ 1 callersFunctiongenerate_warped_noise_from_video
Generate warped noise from a video using make_warped_noise.py. Args: video_path: Path to input video output_dir: Directory t
inference/cogvideox_fun/inference_with_pass1_warped_noise.py:82
↓ 1 callersFunctionget_all_sequences
Get all sequence directories.
data_generation/convert_all_scenarios_for_characters.py:29
↓ 1 callersMethodget_batch
(self, idx)
videox_fun/data/dataset_image_video.py:911
↓ 1 callersMethodget_batch
(self, idx)
videox_fun/data/dataset_image_video_warped.py:936
↓ 1 callersMethodget_batch
(self, idx)
videox_fun/data/dataset_video.py:184
↓ 1 callersFunctionget_bezier_curve
given an array of points *a*, create a curve through those points. *rad* is a number between 0 and 1 to steer the distance of cont
videox_fun/data/dataset_image_video.py:78
↓ 1 callersFunctionget_bezier_curve
given an array of points *a*, create a curve through those points. *rad* is a number between 0 and 1 to steer the distance of cont
videox_fun/data/dataset_image_video_warped.py:78
↓ 1 callersFunctionget_color_for_object
Get realistic color for an object. Args: obj_name: Name of the object character_name: Name of character ('sophie' or 'remy')
data_generation/object_texture_mapping.py:322
↓ 1 callersMethodget_current_frame
Get frame at current_frame_idx from current video
VLM-MASK-REASONER/point_selector_gui.py:304
↓ 1 callersFunctionget_curve
(points, **kw)
videox_fun/data/dataset_image_video.py:63
↓ 1 callersFunctionget_curve
(points, **kw)
videox_fun/data/dataset_image_video_warped.py:63
↓ 1 callersFunctionget_data_config
()
config/default_omnimatte.py:25
↓ 1 callersFunctionget_data_config
()
config/quadmask_cogvideox.py:24
↓ 1 callersFunctionget_data_config
()
config/default_cogvideox.py:24
↓ 1 callersFunctionget_experiment_config
()
config/default_omnimatte.py:13
↓ 1 callersFunctionget_experiment_config
()
config/quadmask_cogvideox.py:13
↓ 1 callersFunctionget_experiment_config
()
config/default_cogvideox.py:13
↓ 1 callersFunctionget_length_to_frame_num
(token_length)
scripts/cogvideox_fun/train.py:1099
↓ 1 callersFunctionget_length_to_frame_num
(token_length)
scripts/cogvideox_fun/train_warped_noise.py:1117
↓ 1 callersFunctionget_light_colors
Get a list of bright, vibrant colors for objects. These colors are distinct from dark human colors (dark grey, dark blue, dark purple). Expand
data_generation/blender_utils.py:270
↓ 1 callersFunctionget_object_physics_type
Get physics type for a specific object in a sequence. Returns: 'active': object should move/fall 'passive': object should st
data_generation/apply_physics_config.py:35
↓ 1 callersFunctionget_omnimatte_config
()
config/default_omnimatte.py:36
↓ 1 callersFunctionget_random_camera_motion
Randomly select a camera motion type. Args: seed: Random seed for reproducible selection Returns: str: One of 'static',
data_generation/blender_utils.py:410
↓ 1 callersFunctionget_random_mask_multi
(shape, mask_type_probs, range_num_masks=[1, 7])
videox_fun/data/dataset_image_video.py:264
↓ 1 callersFunctionget_random_mask_multi
(shape, mask_type_probs, range_num_masks=[1, 7])
videox_fun/data/dataset_image_video_warped.py:264
↓ 1 callersFunctionget_random_points
create n random points in the unit square, which are *mindst* apart, then scale them.
videox_fun/data/dataset_image_video.py:102
↓ 1 callersFunctionget_random_points
create n random points in the unit square, which are *mindst* apart, then scale them.
videox_fun/data/dataset_image_video_warped.py:102
↓ 1 callersFunctionget_resize_crop_region_for_grid
(src, tgt_width, tgt_height)
videox_fun/pipeline/pipeline_cogvideox_fun.py:142
↓ 1 callersMethodget_rotary_embedding
(self, n, device)
videox_fun/reward/MPS/trainer/models/cross_modeling.py:102
↓ 1 callersFunctionget_system_config
()
config/default_omnimatte.py:87
↓ 1 callersFunctionget_system_config
()
config/quadmask_cogvideox.py:58
↓ 1 callersFunctionget_system_config
()
config/default_cogvideox.py:58
↓ 1 callersFunctionget_teacache_coefficients
(model_name)
videox_fun/models/cache_utils.py:5
↓ 1 callersFunctionget_texture_for_object
Get appropriate texture path for an object, or None if it should use color. Args: obj_name: Name of the object Returns:
data_generation/object_texture_mapping.py:283
↓ 1 callersMethodget_timesteps
(self, num_inference_steps, strength, device)
videox_fun/pipeline/pipeline_cogvideox_fun_inpaint.py:729
↓ 1 callersFunctionget_video_dimensions
Get video width and height using ffprobe.
data_generation/convert_masks_to_grid_hybrid.py:21
↓ 1 callersFunctionget_video_mask_validation
( input_video_name, sample_size, max_video_length=49, temporal_window_size=49,
videox_fun/utils/utils.py:418
↓ 1 callersFunctionget_video_model_config
()
config/quadmask_cogvideox.py:35
↓ 1 callersFunctionget_video_model_config
()
config/default_cogvideox.py:35
↓ 1 callersFunctiongrid_cells_to_mask
Convert grid cells to mask
VLM-MASK-REASONER/stage3a_generate_grey_masks.py:160
↓ 1 callersFunctiongridify_mask
Convert pixel mask to gridified mask
VLM-MASK-REASONER/stage3a_generate_grey_masks.py:138
↓ 1 callersFunctiongridify_masks
Gridify masks: if ANY pixel in grid cell → ENTIRE cell = True Args: masks: List of boolean masks (one per frame) grid_rows,
VLM-MASK-REASONER/stage3a_generate_grey_masks_v2.py:164
↓ 1 callersFunctionlinear_decay
(initial_value, final_value, total_steps, current_step)
scripts/cogvideox_fun/train.py:273
↓ 1 callersFunctionlinear_decay
(initial_value, final_value, total_steps, current_step)
scripts/cogvideox_fun/train_warped_noise.py:274
↓ 1 callersFunctionload_and_resize_warped_noise
Load warped noise and resize to match latent dimensions. Args: noise_path: Path to warped noise .npy file target_shape: (lat
inference/cogvideox_fun/inference_with_pass1_warped_noise.py:154
← previousnext →201–300 of 673, ranked by callers