Liang-Chieh Chen

30

Papers

212

Total Citations

Papers (30)

Naive-Student: Leveraging Semi-Supervised Learning in Video Sequences for Urban Scene Segmentation

MaskBit: Embedding-free Image Generation via Bit Tokens

COCONut: Modernizing COCO Segmentation

Semantic Image Segmentation With Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform

MaskLab: Instance Segmentation by Refining Object Detection With Semantic and Direction Features

MobileNetV2: Inverted Residuals and Linear Bottlenecks

Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation

FEELVOS: Fast End-To-End Embedding Learning for Video Object Segmentation

Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation

VIP-DeepLab: Learning Visual Perception With Depth-Aware Video Panoptic Segmentation

DetectoRS: Detecting Objects With Recursive Feature Pyramid and Switchable Atrous Convolution

MaX-DeepLab: End-to-End Panoptic Segmentation With Mask Transformers

CMT-DeepLab: Clustering Mask Transformers for Panoptic Segmentation

TubeFormer-DeepLab: Video Mask Transformer

Weakly- and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation

Searching for MobileNetV3

SPGNet: Semantic Prediction Guidance for Scene Parsing

SegSort: Segmentation by Discriminative Sorting of Segments

Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation

View-Invariant Probabilistic Embedding for Human Pose

Waymo Open Dataset: Panoramic Video Panoptic Segmentation

k-Means Mask Transformer

ViCaS: A Dataset for Combining Holistic and Pixel-level Video Understanding using Captions with Grounded Segmentation

ViTamin: Designing Scalable Vision Models in the Vision-Language Era

Attention to Scale: Scale-Aware Semantic Image Segmentation

Searching for Efficient Multi-Scale Architectures for Dense Image Prediction

Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP

DaTaSeg: Taming a Universal Multi-Dataset Multi-Task Segmentation Model

ReMaX: Relaxing for Better Training on Efficient Panoptic Segmentation

Learning Deep Structured Models