TaskPrompter: Spatial-Channel Multi-Task Prompting for Dense Scene Understanding

Hanrong Ye, Dan Xu
2/12/2026

No official code found. Best implementation: prismformore/Multi-Task-Transformer (326 stars, MIT, last commit 2024-04-24).

Semantic Scholar

Code Implementations(1)

prismformore/Multi-Task-TransformerCommunity61%
MIT

Code of ICLR2023 paper "TaskPrompter: Spatial-Channel Multi-Task Prompting for Dense Scene Understanding" and ECCV2022 paper "Inverted Pyramid Multi-task Transformer for Dense Scene Understanding"

32625C++, CudaJul 4, 20222 years agoMIT
cityscapescityscapes-3dcomputer-visiondeep-learningdepth-estimation+7 more

Similar Papers

HiRA: Parameter-Efficient Hadamard High-Rank Adaptation for Large Language Models

100% match2026

FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation

100% match2026

DriveDreamer: Towards Real-World-Drive World Models for Autonomous Driving

100% match2 repos5812026

Mobile Attention: Mobile-Friendly Linear-Attention for Vision Transformers

100% match1 repo92026

IVTP: Instruction-Guided Visual Token Pruning for Large Vision-Language Models

100% match2026

3DIS: Depth-Driven Decoupled Image Synthesis for Universal Multi-Instance Generation

100% match2026

Cite this paper

@article{ye2026taskprompter,
  title  = {TaskPrompter: Spatial-Channel Multi-Task Prompting for Dense Scene Understanding},
  author = {Hanrong Ye and Dan Xu},
  year   = {2026},
  url    = {https://api.semanticscholar.org/CorpusID:2953eba822d9d65f754f18076977e42781e988b9},
  journal = {ICLR 2023 2023}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2953eba822d9d65f754f18076977e42781e988b9.svg)](https://codeofpaper.com/paper/2953eba822d9d65f754f18076977e42781e988b9)

Discussion