All in Tokens: Unifying Output Space of Visual Tasks via Soft Token
Preprint
Abstract
Unlike language tasks, where the output space is usually limited to a set of tokens, the output space of visual tasks is more complicated, making it difficult to build a unified visual model for various visual tasks. In this paper, we seek to unify the output space of visual tasks, so that we can also build a unified model for visual tasks. To this end, we demonstrate a single unified model that simultaneously handles two typical visual tasks of instance segmentation and depth estimation, which have discrete/fixed-length and continuous/varied-length outputs, respectively. We propose several new techniques that take into account the particularity of visual tasks: 1) Soft token. We employ soft token to represent the task output. Unlike hard tokens in the common VQ-VAE which are assigned one-hot to discrete codebooks/vocabularies, the soft token is assigned softly to the codebook embeddings. Soft token can improve the accuracy of both the next token inference and decoding of the task output; 2) Mask augmentation. Many visual tasks have corruption, undefined or invalid values in label annotations, i.e., occluded area of depth maps. We show that a mask augmentation technique can greatly benefit these tasks. With these new techniques and other designs, we show that the proposed general-purpose task-solver can perform both instance segmentation and depth estimation well. Particularly, we achieve 0.279 RMSE on the specific task of NYUv2 depth estimation, setting a new record on this benchmark. The general-purpose task-solver, dubbed AiT, is available at \url{https://github.com/SwinTransformer/AiT}.Keywords
All Related Versions
This publication has 16 references indexed in Scilit:
- DepthFormer: Exploiting Long-range Correlation and Local Information for Accurate Monocular Depth EstimationMachine Intelligence Research, 2023
- Inverted Pyramid Multi-task Transformer for Dense Scene Understanding2022
- Neural Window Fully-connected CRFs for Monocular Depth Estimation2022
- P3Depth: Monocular Depth Estimation with a Piecewise Planarity Prior2022
- SDC-Depth: Semantic Divide-and-Conquer Network for Monocular Depth Estimation2020
- Monocular Depth Estimation Using Relative Depth Maps2019
- Structured Attention Guided Convolutional Neural Fields for Monocular Depth Estimation2018
- GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation2018
- Deep Ordinal Regression Network for Monocular Depth Estimation2018
- Deep convolutional neural fields for depth estimation from a single image2015