Open Models Almanac
Methodsmodel architecturemultimodal architecture

single mention · model architecture

Hierarchical convolutional decoder

A hierarchical convolutional decoder with a pyramid structure composed of a shared convolutional neck and multiple task-specific heads.1 The neck progressively upsamples features into a multi-scale feature pyramid that is shared across all task heads, so each head decodes its own dense output while features are reused.2

How this definition was written

Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.

  1. The decoder follows a pyramid structure composed of a shared convolutional neck and multiple task-specific heads — Masked Depth Modeling for Spatial Perception
  2. The resulting multi-scale feature pyramid is shared across all task heads, enabling efficient feature reuse while allowing each head to decode its own dense output — Masked Depth Modeling for Spatial Perception

Filed under model architecture › multimodal architecture