《多媒体知识点答案——部分.doc》由会员分享，可在线阅读，更多相关《多媒体知识点答案——部分.doc（5页珍藏版）》请在金锄头文库上搜索。
1、1. Explain the meaning of FOURCC tags in AVI container “00db”, “01dc”, “01wb”, “vids”, “auds”, etc. 2. What is audio CD data rate? 44.1 kHz * 16 *2 = 1.4112 Mb/s3. What are audio coding goals? Fidelity: Maximize the perceived audio quality at the audio decoder (minimize the distortion). Data rate: M
2、inimize the amount of digital data to represent the original audio signal. Complexity: Minimize the computational complexity. Delay: Minimize the coding delay.4. Explain the structure of Waveformat Chunk Typedef struct ID chunkID; long chunkSize; short wFormatTag; unsigned short wChannels; unsigned
3、long dwSamplesPerSec; unsigned long dwAvgBytesPerSec; unsigned short wBlockAlign; unsigned short wBitsPerSample; unsigned short cbSize*; FormatChunk;5. Describe potential coding errors in audio coding. Sampling errorsAliasing effect; Quantization errors: Overload errors Round-off errors Storage and
4、transmission errors;6. Explain the difference between lossy compression and lossless compression. 所谓有损压缩是利用了人类对图像或声波中的某些频率成分不敏感的特性，允许压缩过程中损失一定的信息；无损压缩是利用统计冗余进行编码压缩，可完全恢复原始数据而不引起任何失真。7. Describe the difference between uniform quantization and non-uniform quantization. 把输入信号的取值域等间隔分割的量化称为均匀量化。非均匀量化是一种
5、在输入信号的动态范围内量化间隔不相等的量化。8. Describe the Bitmap Structure public class BitmapInfoHeader public int biSize; public int biWidth; public int biHeight; public short biPlanes; public short biBitCount; public int biCompression; public int biSizeImage; public int biXPelsPerMeter; public int biYPelsPerMeter; p
6、ublic int biClrUsed; public int biClrImportant; 9.What is the definition of entropy?信源的平均不确定度。10. Describe the procedure of Huffman coding.Huffman树的构建；11. Describe ADPCM coding process and the full name of ADPCM in EnglishADPCM ：Adaptive Differential Pulse Code Modulation12. What is 010 editor? How
7、to define struct and array?13. Describe the difference among Fourier transform, Fourier series, Discrete time Fourier transform and Discrete Fourier transform.傅里叶变换在时域和频域上都是连续的；傅里叶级数时域有限，频域上是离散；离散时间傅里叶变换时域上离散的，频域上是周期或离散的；离散傅里叶变换在时域和频域上都是离散的。14. What is the computation complexity of decimation in tim
8、e FFT?N*LogN。15. What is the purpose of windowing before time to frequency mapping?使得无限时长的信号截取为有限长的信号，便于计算机处理。16. What are four types of redundancy in image and video coding?17. In image and video compression why do we prefer using YUV color space rather than RGB space?应用了视觉冗余，这样便可以对色度分量使用更少的采样点。18.
9、 What is sampling resolution for CIF?抽样分辨率。19. What are three basic types of coded frame in video coding?I，P，B20. Explain motion estimation and compensation将活动图像分成若干块或宏块,并搜索出每个宏块在邻近帧图像中的位置,计算出两者之间的空间位置的相对偏移量,那么得到的相对偏移量就是通常所指的运动矢量,而得到运动矢量的过程称为运动估计. 当运动矢量为分数像素精度的时候，预测块并不真实存在，需要通过内插的方式得到。因此得到了运动矢量并不等于就
10、得到了预测块。所以运动补偿就是利用运动矢量得到预测块的过程。21. What is PSNR?峰值信噪比。22. What is the purpose of DCT?减少数据相关性，便于后续熵编码和其他处理。23. Why is zigzag scan applied before entropy coding in JPEG?使得更多的零系数放在一起，便于熵编码。24. How to address the issue of bit error transmission for JPEG compressed bit streams?插入同步字节。25. Why do we apply d
11、ifferential coding to DC coefficients in the process of JPEG compression?DC系数变化比较小，因此使用差分编码。26. What is the difference between an MCU and a data unit? Describe their relationship.Minimum Coded Unit,27. Describe four modes of operation employed in JPEG compression.based, progressive DCT-based, lossle
12、ss, and hierarchical. 28. In JPEG AC entropy coding, what does run length mean? What is the maximum value of run? What does size mean? What does a (0, 0) run size pair mean? 0系数的长度；非零系数的最大值；非零系数的最大值需要用多少位来表示；（EOB）（0，0）代表结束29. What is the difference between VLC and VLI in JPEG coding? VLCs and VLIs a
13、re codes with variable lengths, but VLIs are not Huffman codes.VLC 和VLI均是变长码，但是VLI不是Huffman码，它的编码分为两部分，前半部分为该符号所占位数，后半部分才是实际值。30. What are markers used for in JIF? SOI, EOI , SOF, SOS31. Describe the full name of SOI, EOI , SOF, SOS.32. What is JPEG file interchange format used for? 用于JPEG标准的数据存储。33
14、. Describe JPEG encoding and decoding procedure respectively. 34. Describe the applications of H.261, H.263, MPEG-2H.261 视频会议，H.263 视频电话， MPEG-2 DVD35. Describe video encoding and decoding procedure.H.261使用了混合编码框架，包括了基于运动补偿的帧间预测，基于离散余弦变换的空域变换编码，量化，zig-zag扫描和熵编码。36. What is GOP? What is a video seque
15、nce? What is a slice?Group Of Picture，序列包含多个GOP，多个slice组成一个GOP。37. What is H.264 MV resolution? Motion Vector 分辨率1/4 Pixel38. Does H.261 allow B frame?No39. Which video coding method uses 4x4 DCT transform? H.264 wmv940. Is the video data stored in coding order or display order? Coding order41. Does video coding standard specify encoding process or specific decoder implementation? No42. Which of the following is tru
