Animate camera along trajectory.

r""" Returns: `dict` of attention processors: A dictionary containing all attention processors used in the model with

Clear all points for current video

Clear points for current frame only

Clear drawn trajectory

Fully batched computation of point penetration using Kaolin's optimized operations. Note: Kaolin's check_sign doesn't suppor

Copy ONLY black component from previous frame, preserving grey in current frame

Copy ONLY grey component from previous frame, preserving black in current frame

Create Blender armature from bone hierarchy.

Create Blender object from vertices and faces.

r""" Disable sliced VAE decoding. If `enable_slicing` was previously enabled, this method will go back to computing decoding in one st

r""" Disable tiled VAE decoding. If `enable_tiling` was previously enabled, this method will go back to computing decoding in one step

r""" Enable sliced VAE decoding. When this option is enabled, the VAE will split the input tensor in slices to compute decoding in sev

r""" Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to compute decoding and en

Standalone function to extract VideoJAM motion from a single video file. IMPORTANT: For training, set target_resolution to match your traini

Go to first frame

Jump to first frame

Jump to first video

Forward pass to compute deformed vertices and joint positions Args: pose_params: Dict of bone_name -> tensor [batch_size,

einstein notation b - batch h - heads n, i, j - sequence length (base sequence length, source, target) d - fe

einstein notation b - batch h - heads n, i, j - sequence length (base sequence length, source, target) d - fe

r"""Forward method of the `CogVideoXDownBlock3D` class.

r"""Forward method of the `CogVideoXMidBlock3D` class.

r"""The forward method of the `CogVideoXEncoder3D` class.

r"""The forward method of the `CogVideoXDecoder3D` class.

r""" Args: text_embeds (`torch.Tensor`): Input text embeddings. Expected shape: (batch_size, seq_length, embedding

Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att

r"""Enables fused QKV projections.

r"""Enables fused QKV projections.

Generate smooth linear camera movement (left or right).

Get the chain of joints between two bones for IK Args: start_bone: Starting bone name end_bone: Ending bone n

Get the possible texture categories for an object. Args: obj_name: Name of the object Returns: list: List of texture ca