Computer Vision


Agriculture x AI

Chick Sexing on Face Part

AI Processing Plant

  • “Chicken Processing Plant With Automated Computer Vision”
  • “Artificial Intelligence And Vision-Based Broiler Body Weight Measurement System And Process”





Geospatial x AI

Aerial Image Segmentation

  • “AerialFormer: Multi-resolution Transformer for Aerial Image Segmentation”

Solar PV Profiling

  • “SolarFormer: Multi-scale Transformer for Solar PV Profiling”





Video Understanding

Video Anomaly Detection

  • “CLIP-TSA: CLIP-Assisted Temporal Self-Attention for Weakly-Supervised Video Anomaly Detection” (ICIP 2023)

Video Paragraph Captioning

  • “VLTinT: Visual-Linguistic Transformer-in-Transformer for Coherent Video Paragraph Captioning” (AAAI 2023)
  • “Vlcap: Vision-language with contrastive learning for coherent video paragraph captioning” (ICIP 2022)

Temporal Action Proposal Generation

  • “Aoe-net: Entities interactions modeling with adaptive attention mechanism for temporal action proposals generation” (IJCV)
  • “AEI: Actors-Environment Interaction with Adaptive Attention for Temporal Action Proposals Generation” (BMVC 2021)

Image Understanding

Amodal Image Segmentation

  • “AISFormer: Amodal Instance Segmentation with Transformer” (BMVC 2022)





Robotics


Perception

Queryable Scene Reconstruction

  • “Open-Fusion: Real-time Open-Vocabulary 3D Mapping and Queryable Scene Representation” (ICRA 2024)

Deformable Human 3D Reconstruction

  • “DNA: Deformable Neural Articulations Network for Template-Free Dynamic 3D Human Reconstruction From Monocular RGB-D Video” (CVPRW 2023)


Locomotion

Robust Gait Learning

While quadrupeds can open the operational domains of robots thanks to their dynamic locomotion capabilities, conventional controllers for legged locomotion constraint their applications to relatively simple environments that can be taken over by wheeled robots. Here we use reinforcement learning to train a quadruped to walk on various terrains. In the simulation, a quadruped robot (Unitree Go1) learns to walk across challenging terrain, including uneven surfaces, slopes, stairs, and obstacles, while following linear- and angular- velocity commands.


Medical Imaging


Medical Image Segmentation

Volumetric Segmentation

  • “SAM3D: Segment Anything Model in Volumetric Medical Images”
  • “DAM-AL: Dilated Attention Mechanism with Attention Loss for 3D Infant Brain Image Segmentation”
  • “Point-unet: A context-aware point-based neural network for volumetric segmentation”
  • “Invertible residual network with regularization for effective volumetric segmentation” (SPIE 2021)

Capsule Networks

  • “3DConvCaps: 3DUnet with Convolutional Capsule Encoder for Medical Image Segmentation”
  • “CapsNet for Medical Image Segmentation”
  • “SS-3DCapsNet: Self-supervised 3D Capsule Networks for Medical Segmentation on Less Labeled Data” (ISBI 2022)
  • “3d-ucaps: 3d capsules unet for volumetric image segmentation”

Interpretable AI System

CXR Diagnoses

  • “I-AI: A Controllable & Interpretable AI System for Decoding Radiologists’ Intense Focus for Accurate CXR Diagnoses” (WACV 2024)