paper-with-me

Papers

Semi-Autoregressive Transformer for Image Captioning

2021-06-17 · Yuanen Zhou, Yong Zhang, Zhenzhen Hu, Meng Wang

Current state-of-the-art image captioning models adopt autoregressive decoders, \ie they generate each word by conditioning on previously generated words, which leads to heavy latency during inference. To tackle this issue, non-autoregressive image captioning models have recently been proposed to significantly accelerate the speed of inference by generating all words in parallel. However, these non-autoregressive models inevitably suffer from large generation quality degradation since they remove words dependence excessively. To make a better trade-off between speed and quality, we introduce a semi-autoregressive model for image captioning~(dubbed as SATIC), which keeps the autoregressive property in global but generates words parallelly in local . Based on Transformer, there are only a few modifications needed to implement SATIC. Experimental results on the MSCOCO image captioning benchmark show that SATIC can achieve a good trade-off without bells and whistles. Code is available at {\color{magenta}\url{https://github.com/YuanEZhou/satic}}.

📄 PDF Abstract BibTeX arXiv:2106.09436

Code (1)

YuanEZhou/satic 공식 구현 pytorch

Tasks

Image Captioning

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Position-Wise Feed-Forward Layer 설명 없음
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Adam 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Semi-Autoregressive Image Captioning

2021-10-11 · Xu Yan, Zhengcong Fei, Zekang Li, Shuhui Wang 외

Current state-of-the-art approaches for image captioning typically adopt an autoregressive manner, i.e., generating descriptions word by word, which suffers from slow decoding issue and becomes a bottleneck in real-time …

DecoderImage CaptioningSentence

Bounding and Filling: A Fast and Flexible Framework for Image Captioning

2023-10-15 · Zheng Ma, Changxin Wang, Bo Huang, Zixuan Zhu 외

Most image captioning models following an autoregressive manner suffer from significant inference latency. Several models adopted a non-autoregressive manner to speed up the process. However, the vanilla non-autoregressi…

Image CaptioningImage Description

Non-Autoregressive Image Captioning with Counterfactuals-Critical Multi-Agent Learning

2020-05-10 · Longteng Guo, Jing Liu, Xinxin Zhu, Xingjian He 외

Most image captioning models are autoregressive, i.e. they generate each word by conditioning on previously generated words, which leads to heavy latency during inference. Recently, non-autoregressive decoding has been p…

Image CaptioningMachine TranslationMulti-agent Reinforcement LearningSentence+1

Semantic-Conditional Diffusion Networks for Image Captioning

2022-12-06 · CVPR 2023 1 · Jianjie Luo, Yehao Li, Yingwei Pan, Ting Yao 외

Recent advances on text-to-image generation have witnessed the rise of diffusion models which act as powerful generative models. Nevertheless, it is not trivial to exploit such latent variable models to capture the depen…

Cross-Modal RetrievalDecoderImage CaptioningImage Generation+4

Length-Controllable Image Captioning

2020-07-19 · ECCV 2020 8 · Chaorui Deng, Ning Ding, Mingkui Tan, Qi Wu

The last decade has witnessed remarkable progress in the image captioning task; however, most existing methods cannot control their captions, \emph{e.g.}, choosing to describe the image either roughly or in detail. In th…

controllable image captioningDecoderDiversityImage Captioning