h->predict_16x16[i_mode]( p_dst, i_stride ); 计算对应预测模式时的预测采样值。输出放到dst指向的数组中。Pred0ct_16x16是7个元素指向的数组，数组的每个元素是一个指向函数的指针变量，在x264_predict_16x16_init函数初始这个指针数组。7个元素分别对应16X16的帧内预测时不同的预测模式。分别是水平，垂直，PLANE，DC和

【论文笔记】An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale(Vision Transformer, ViT) 文章题目：An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale作者：Dosovitskiy

一幅图像相当于16X16个words:大规模图像识别的Transformer paper题目：AN IMAGE IS WORTH 16X16 WORDS: TRANSFORMERS FOR IMAGE RECOGNITION AT SCALE paper是Google发表在ICLR 2021的工作 paper地址：链接 ABSTRACT 虽然Transformer体系结

“We show that this reliance on CNNs is not necessary and a pure transformer applied directly to sequences of image patches can perform very well on image classification tasks.” ——完全不依赖CNN 参考：Vision T