摘要:
这是一篇 text to 3D 方向的突破性的文章,效果确实非常棒,据说一作的朋友圈中也说“他们将这个领域从20分提升到了70分的水平”,预测之后会有许多基于该方法的优秀文章与产品出现,毕竟之前 SDS follow 的文章也有很大一批。本阅读笔记就简要去记录一下这篇论文的主要方法和思想。 ![im 阅读全文 »
摘要:
Introduction Previous Work bundle adjustment or loop closure VoxHashing with Signed Distance Function deep geometry learning Challenges explicitly mod 阅读全文 »
摘要:
Various Sequence To Sequence Architectures Basic Models Sequence to sequence model Image captioning use CNN(AlexNet) first to get a 4096-dimensional v 阅读全文 »
摘要:
Word Embeddings Word Representation 1-hot representation: any product of them is 0 Featurized representation: word embedding Visualizing word embe 阅读全文 »