Jiachen Lu

7

Papers

28

Total Citations

Papers (7)

Enhancing High-Resolution 3D Generation through Pixel-wise Gradient Clipping

LaneGraph2Seq: Lane Topology Extraction with Language Model via Vertex-Edge Encoding and Connectivity Enhancement

Translating Images to Road Network: A Non-Autoregressive Sequence-to-Sequence Approach

Learning Ego 3D Representation As Ray Tracing

Generative Semantic Segmentation

Rethinking Semantic Segmentation From a Sequence-to-Sequence Perspective With Transformers

SOFT: Softmax-free Transformer with Linear Complexity