https://github.com/MindCode-4/code-6/tree/main/drop-an-octave-reducing-spatial
Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Image Classification is a fundamental task in vision recognition that aims to understand and categorize an image as a whole under a specific label. Unlike object detection, which involves classification and location of multiple objects within an image, image classification typically pertains to single-object images. When the classification becomes highly detailed or reaches instance-level, it is often referred to as image retrieval, which also involves finding similar images in a large database.
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution
DropBlock: A regularization method for convolutional networks
Fixup Initialization: Residual Learning Without Normalization
Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels
FALFormer: Feature-aware Landmarks self-attention for Whole-slide Image Classification
Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solution
PolyLoss: A Polynomial Expansion Perspective of Classification Loss Functions
ResMLP: Feedforward networks for image classification with data-efficient training
VATT: Transformers for Multimodal Self-Supervised Learning from Raw Video, Audio and Text
Weakly Supervised Learning of Instance Segmentation with Inter-pixel Relations
Improving ProtoNet for Few-Shot Video Object Recognition: Winner of ORBIT Challenge 2022
Very Deep Convolutional Networks for Large-Scale Image Recognition
ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
Video Classification with Channel-Separated Convolutional Networks
TNT: Text-Conditioned Network with Transductive Inference for Few-Shot Video Classification
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
SpineNet: Learning Scale-Permuted Backbone for Recognition and Localization
Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.