hiddenlayers.tech reading list

Perception

SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers (Xie, Enze, et al., 2021)

Panoptic SegFormer: Delving Deeper into Panoptic Segmentation with Transformers (Li et al., 2022)

Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth (Kim et al., 2022)

Deformable DETR: Deformable Transformers for End-to-End Object Detection

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

• Introduces Spatial-Resolution Attention (SRA), the efficient self-attention used in Segformer.

Panoptic Segmentation (Kirillov et al., 2019)

How Much Position Information Do Convolutional Neural Networks Encode? (Islam et al., 2020)

• Shows that CNNs actually do encode positional information that seems to be provided by zero padding.

Prediction, Planning, Control

DiffStack: A Differentiable and Modular Control Stack for Autonomous Vehicles (Karkus et al., 2023)

Trajectron++: Dynamically-Feasible Trajectory Forecasting With Heterogeneous Data (Salzmann et al., 2020)