You Only Learn One Representation: Unified Network for Multiple Tasks

Abstract

People ``understand'' the world via vision, hearing, tactile, and also the past experience. Human experience can be learned through normal learning (we call it explicit knowledge), or subconsciously (we call it implicit knowledge). These experiences learned through normal learning or subconsciously will be encoded and stored in the brain. Using these abundant experience as a huge database, human beings can effectively process data, even they were unseen beforehand. In this paper, we propose a unified network to encode implicit knowledge and explicit knowledge together, just like the human brain can learn knowledge from normal learning as well as subconsciousness learning. The unified network can generate a unified representation to simultaneously serve various tasks. We can perform kernel space alignment, prediction refinement, and multi-task learning in a convolutional neural network. The results demonstrate that when implicit knowledge is introduced into the neural network, it benefits the performance of all tasks. We further analyze the implicit representation learnt from the proposed unified network, and it shows great capability on catching the physical meaning of different tasks. The source code of this work is at : https://github.com/WongKinYiu/yolor.

References in corpus (5)

• Deep Equilibrium Models
• Probabilistic two-stage detection
• Scaled-YOLOv4: Scaling Cross Stage Partial Network
• Selective Kernel Networks
• Implicit Feature Pyramid Network for Object Detection

Cited by in corpus (10)

• Swin Transformer V2: Scaling Up Capacity and Resolution
• Automated Defect Detection for Mass-Produced Electronic Components Based on YOLO Object Detection Models
• Who Is Alyx? A new Behavioral Biometric Dataset for User Identification in XR
• Tracking perovskite crystallization via deep learning-based feature detection on 2D X-ray scattering data
• Identification of Binary Neutron Star Mergers in Gravitational-Wave Data Using YOLO One-Shot Object Detection
• Online V2X Scheduling for Raw-Level Cooperative Perception
• Table Detection for Visually Rich Document Images
• Image-Based Vehicle Classification by Synergizing Features from Supervised and Self-Supervised Learning Paradigms
• 2nd Place Solution for Waymo Open Dataset Challenge -- Real-time 2D Object Detection
• Methodology for generating synthetic labeled datasets for visual container inspection