An attention layer that allows for downscaling the size of the embedding after projection to queries, keys, and values.

Visualizer that draws data about detection/segmentation on images. It contains methods like `draw_{text,box,circle,line,binary_mask,polygon}

A structure for storing masks and their related data in batched format. Implements basic filtering and concatenation.

Attribute: polygons (list[ndarray]): list[ndarray]: polygons for this mask. Each ndarray has format [x, y, x, y, ...]

Attention with rotary position encoding.

A list of video frames to be load asynchronously without blocking session start.

This class represents a batch of videos with associated annotations. Attributes: img_batch: A [TxBxCxHxW] tensor containing the image

r"""ConvNeXt Block. There are two equivalent implementations: (1) DwConv -> LayerNorm (channels_first) -> 1x1 Conv -> GELU -> 1x1 Conv; all in (N,

Image to Patch Embedding.

Positional encoding using random frequencies for 1D inputs.

Positional encoding using random spatial frequencies.

Enum of different color modes to use for instance visualizations.

A modified variant of Feature Pyramid Network (FPN) neck (we remove output conv and also do bicubic interpolation similar to ViT pos embe

Reference: https://arxiv.org/abs/2306.00989

Progressively downsample a mask by total_stride, each time by stride. Note that LayerNorm is applied per *token*, like in ViT. With each

This is a more standard version of the position embedding, very similar to the one used by the Attention Is All You Need paper, generalized t

The predictor class to handle user interactions and manage inference states.

Optimized for the VOS setting