mem_k: [B, size, C] mem_v: [B, size, C] feat_k: [B, 1, C] feat_v: [B, 1, C]

Init. Args: scale_factor (float): scaling mode (str): interpolation mode

Current Version: input the first frame img feature and pose to initialize the state feature and pose

Inverse to pose_encoding_to_camera camera: opencv, cam2world

Read depth data from file, return as numpy array.

Args: - depthmap (HxW array): - camera_intrinsics: a 3x3 matrix - camera_pose: a 4x3 or 4x4 cam2world matrix Returns:

Args: - depthmap (BxHxW array): - pseudo_focal: [B,H,W] ; [B,2,H,W] or [B,1,H,W] Returns: pointmap of absolute coordi

Convert an image basename to an integer timestamp. For ScanNet data, it is assumed that the basename is an integer string. Args:

tqdm but with parallel execution. Will essentially return res = [ function(arg) # default function(*arg) # if star_args is Tr

extract 3D points from prediction head output

Removes the 'module.' prefix from the keys of a state_dict. Args: state_dict (dict): The original state_dict with possible 'module.'

Transfer some variables to another device (i.e. GPU, CPU:torch, CPU:numpy). batch: list, tuple, dict of tensors or other things device: pytor

Transfer some variables to another device (i.e. GPU, CPU:torch, CPU:numpy). batch: list, tuple, dict of tensors or other things device: pytor

Output a (H,W,2) array of int32 with output[j,i,0] = i + origin[0] output[j,i,1] = j + origin[1]

Return a color jitter transform with same random parameters

flow_utils._computeColor(flow): compute color codes for the flow field flow ---- PARAMETERS ---- flow: np.array of dimension (height

return_all_blocks: if True, return the features at the end of every block instead of just the features from the la

image has B x 3 x img_size x img_size do_mask: whether to perform masking or not return_all_blocks: if True, return the featu

Convert COLMAP intrinsics (with pixel centers at (0.5, 0.5)) to OpenCV convention.

Modify camera intrinsics to follow a different convention. Coordinates of the center of the top-left pixels are by default: - (0.5, 0.5)

turn a grayscale image into a color image :param x: torch.Tensor, grayscale image, [H, W] or [B, H, W] :param mask: torch.Tensor or None,

turn a grayscale image into a color image :param x: torch.Tensor, grayscale image, [H, W] or [B, H, W] :param mask: torch.Tensor or None,

Compute 'overlapping' metrics based on a distance threshold between two point clouds.

gt_pose: list of (Bx3, Bx4) pred_pose: list of (Bx3, Bx4) masks: None, or B

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [n_cls_token+grid_size*grid_size, embe