https://github.com/souvikshanku/show-attend-and-tell
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
Image Captioning is the task of describing the content of an image in words. This task lies at the intersection of computer vision and natural language processing. Most image captioning systems use an encoder-decoder framework, where an input image is encoded into an intermediate representation of the information in the image, and then decoded into a descriptive text sequence. The most popular benchmarks are nocaps and COCO, and models are typically evaluated according to a BLEU or CIDER metric.
( Image credit: Reflective Decoding Network for Image Captioning, ICCV'19)
관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
A Hierarchical Approach for Generating Descriptive Image Paragraphs
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models
Show and Tell: Lessons learned from the 2015 MSCOCO Image Captioning Challenge
Visual Grounding Helps Learn Word Meanings in Low-Data Regimes
MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition
Show and Tell: Lessons learned from the 2015 MSCOCO Image Captioning Challenge
Scheduled Sampling for Sequence Prediction with Recurrent Neural Networks
Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering
Semantic Object Accuracy for Generative Text-to-Image Synthesis
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Features
ReConPatch : Contrastive Patch Representation Learning for Industrial Anomaly Detection
RISE: Randomized Input Sampling for Explanation of Black-box Models
A Comprehensive Survey of Deep Learning for Image Captioning
Text-Only Training for Image Captioning using Noise-Injected CLIP
Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.