-
非常感谢您开源代码，我仔细阅读了您的文章并重新复现了代码，收获很多。但是也存在一些疑问：
提取300个单词作为语义分支的训练标签，提取语义特征，该特征被映射到0-1之间。
问题1：语义特征和lstm的视觉输入、前一时刻的word-embedding,前一时刻的隐藏层h(t-1)分别对应相乘，达到语义融合的目的吗？
问题2：文章提到人工选取k个单词，那么选取的标准是什么呢？为什么选择300个，…
-
Hello!
From previous issues I have noticed that you have provided two datasets (http://www.zeeshanzia.com/dataset/KITTI3D_cvpr17.zip and http://www.nec-labs.com/~mas/DISCO/). However, I noticed that …
-
Add ImageNet as a benchmark based on the Pytorch data loader.
-
Hi,
First of all awesome work. Your this repository helped me a lot during my literature review and my research work.
Second, just a minor addition. Here is a great paper "Learning Discriminative an…
-
Thought it was worth including this related paper I came across from naverlabs.
"From handcrafted to deep local features", Gabriela Csurka, Christopher R. Dance, Martin Humenberger. 2019
https://a…
-
I found some unrelated papers in this list, such as:
Non-local Neural Networks. Wang, Xiaolong and Girshick, Ross and Gupta, Abhinav and He, Kaiming. CVPR 2018.
Neural Message Passing for Quant…
-
Forwarded from e-mail:
-----
Dear Prof. Berger,
Thank you for establishing, running, and maintaining the csrankings.org website. Your website has become a highly valuable resource for prospec…
-
你好，我用的是COCO数据集，
两层LSTM模型， 一层用于实现自上而下的注意力，一层实现语言模型。
我将所有图片描述中，出现频率大于3次的词作为字典文件，一共有14226个词。
words = [w for w in word_freq.keys() if word_freq[w] > 3]
训练好模型后，在使用时，结果中出现多个同类型的词，比如：
放在 床上 的 笔记 笔记…
-
Thanks for your nice work and code release.
- I wonder if there is any script for the training procedure in Alg.1 of your paper? I followed your paper and wrote my own version (without ground truth)…
-
**Submitting author:** @ablaom (Anthony Blaom)
**Repository:** https://github.com/alan-turing-institute/MLJ.jl
**Version:** v0.14.1
**Editor:** @terrytangyuan
**Reviewer:** @degleris1, @henrykironde
*…