References ▼
This is the interface for the LEAD MPEG-4 Encoder.
Lists the possible encoding threads number that can be set on the property EncodingThreads.
typedef enum{LMP4_THREAD_AUTO,LMP4_THREAD_1,LMP4_THREAD_2,LMP4_THREAD_3,LMP4_THREAD_4,LMP4_THREAD_5,LMP4_THREAD_6,LMP4_THREAD_7,LMP4_THREAD_8,} eEncodingThreads;
Have the LEAD MPEG4 Encoder determine the appropriate number of threads.
Use 1 thread during encoding.
Use 2 threads during encoding.
Use 3 threads during encoding.
Use 4 threads during encoding.
Use 5 threads during encoding.
Use 6 threads during encoding.
Use 7 threads during encoding.
Use 8 threads during encoding.
Lists the possible encoding passes counts.
typedef enum{LMP4_ENCODINGPASSESCOUNT_ONE,LMP4_ENCODINGPASSESCOUNT_TWO,} eEncodingPassesCount;
Use 1 pass mode in encoding.
Use 2 passes mode in encoding.
Lists the possible values for the compression mode.
typedef enum{LMP4_COMPRESSIONMODE_ConstantQuality,LMP4_COMPRESSIONMODE_VariableBitrate,LMP4_COMPRESSIONMODE_ConstantBitrate,} eCompressionMode;
Uses the QFactor property value to regulate the image quality.
Varies the quality over short durations. However, it will achieve an average rate matching the BitRate property value.
Strictly adheres to the rate specified in the BitRate property.
Lists the possible values for the output format.
typedef enum{LMP4_OUTPUTFORMAT_mp4v,LMP4_OUTPUTFORMAT_LMP4,LMP4_OUTPUTFORMAT_XVID,LMP4_OUTPUTFORMAT_iPod,} eOutputFormat;
Make the encoder output type the Standard MPEG4 video media subtype.
Make the encoder output type the LEAD proprietary MPEG4 video subtype. This allows super compression to be used.
Make the encoder output type the XVID video subtype.
Make the encoder output compatible with iPod devices.
Lists the possible values for the quantization method to be used by the encoder.
typedef enum{LMP4_QUANTTYPE_Linear,LMP4_QUANTTYPE_NonLinear,} eQuantType;
Use H.263 quantization.
Use MPEG quantization.
Lists the possible values for the Motion estimation accuracy.
typedef enum{LMP4_MOTIONACCURACY_FullPixel,LMP4_MOTIONACCURACY_HalfPixel,} eMotionAccuracy;
One pixel accuracy.
Half pixel accuracy (faster).
List the possible values for the motion algorithm. (the algorithm to be used when determining if motion is occurring).
typedef enum{LMP4_MOTIONALGORITHM_FullSearch,LMP4_MOTIONALGORITHM_FastSearch,} eMotionAlgorithm;
Use the full motion search algorithm (testing all possibilities). This option slows down the encoding process considerably.
Use fast motion search.
List the possible values for the motion vectors (vectors to be used in the motion estimation algorithm).
typedef enum{LMP4_MOTIONVECTOR_OnePerMacroblock,LMP4_MOTIONVECTOR_FourPerMacroblock,} eMotionVectorPerMacroblock;
Use one motion vector (fast).
Use four motion vectors (better quality, slower encoding).
| Type | Name | Description |
|---|---|---|
| long | QFactor | The quality factor used in compressing the video stream; the lower the quality factor, the better the quality. Possible values range from 1 (best) to 31 (worst). |
| eQuantType | QuantType | The quantization method applied. |
| double | OutputFrameRate | The desired output frame rate, in frames per second. Possible values range from 1.0 to 30.0. Pass -1 to make the output frame rate the same as the input frame rate. |
| eOutputFormat | OutputFormat | The encoder output type. |
| VARIANT_BOOL | SuperCompression | Enables/disables super compression, LEADs improvement to the MPEG4 video compression. When super compression is used, compressed video size is reduced noticeably with almost the same quality, and the encoding and decoding speed is enhanced. Files generated using this option, will not be MPEG4 compliant: they can be decoded only by the LEAD MPEG4 Video Decoder. Possible values are: |
| • VARIANT_TRUE - Enable super compression. This is allowed only if the output format is set to LMP4_OutputFormat_LMP4. | ||
| • VARIANT_FALSE - Disable super compression. | ||
| long | PFrames | Specifies the number of P frames between two successive I frames. * |
| long | BFrames | Specifies the number of B frames between two successive P frames. * |
| long | SearchWidth | The search block width used when determining if motion is occurring. The larger the block, the slower the encoding process. It is not recommended to use large values. The range is from 2 to the video stream width, in pixels. |
| long | SearchHeight | The search block height used when determining if motion is occurring. The larger the block, the slower the encoding process. It is not recommended to use large values. The range is from 2 to the video stream height, in pixels. |
| eMotionAlgorithm | MotionAlgorithm | The algorithm to be used when determining if motion is occurring. |
| eMotionAccuracy | MotionAccuracy | Motion estimation accuracy. |
| eMotionVectorPerMacroblock | MotionVectorPerMacroblock | Specifies how many motion vectors are used in the motion estimation algorithm. |
| eEncodingThreads | EncodingThreads | The number of threads used during encoding. The maximum number of threads that can be used is 8. Using multi threads with a multi core machine or multi CPUs will increase the encoding speed significantly. |
| VARIANT_BOOL | InsertVOL | Set to VARIANT_TRUE to insert a mpeg4 header before each I-frame. This is useful in a video broadcasting where the mpeg4 decoder might start decoding from the middle of the stream. |
| VARIANT_BOOL | InsertGOV | Set to VARIANT_TRUE to insert an optional header before I frames that indicates to the decoder the timestamp of the I frame and whether it is open/closed GOV. This helps the stream to be randomly seeked in some decoders. However, the LEAD MPEG4 decoder does not require it. |
| eCompressionMode | CompressionMode | Sets the compression mode. |
| long | BitRate | This is the average or constant bitrate used when the CompressionMode is set to LMP4_COMPRESSIONMODE_VariableBitrate or LMP4_COMPRESSIONMODE_ConstantBitrate. The value is in bits per second. |
| eMotionAccuracy | MotionAccurancy | Motion estimation accuracy. |
| VARIANT_BOOL | Interlacing | Set to VARIANT_TRUE to encode the video as interlaced fields. Otherwise, the video is encoded as frames. The default is VARIANT_FALSE. |
| long | SceneChangeThreshold | The threshold that determines a scene change point in a video clip. The encoder calculates this value for every inter-frame being encoded. If the calculated value is greater than threshold then it is a scene-change point and the frame will be encoded as an I-frame. this value should be between 1 and 100 and default is 50, (values between 30-50 gives the best compression result). To turn off this feature Scene Change Detection, should be set to 100. |
| eEncodingPassesCount | EncodingPasses | Specifies the number of encoding passes. |
Resets the filter properties to their default values. Calling this method can cause the filter to reconnect its output pin. The default values are:
| Property | Default value |
|---|---|
| OutputFrameRate | -1.0 |
| PFrames | 9 |
| BFrames | 0 |
| OutputFormat | LMP4_OUTPUTFORMAT_mp4v |
| BitRate | 300000 |
| interlaced | VARIANT_FALSE |
| InsertVOL | VARIANT_TRUE |
| insertGOV | VARIANT_FALSE |
| SceneChangeThreshold | 50 |
| EncodingPassesCount | LMP4_ENCODINGPASSESCOUNT_ONE |
| EncodingThreads | LMP4_THREAD_1 |
| CompressionMode | LMP4_COMPRESSIONMODE_ConstantQuality |
| MotionVectorPerMacroblock | LMP4_MOTIONVECTOR_OnePerMacroblock |
| MotionAccuracy | LMP4_MOTIONACCURACY_FullPixel |
| MotionAlgorithm | LMP4_MOTIONALGORITHM_FastSearch |
| QuantType | LMP4_QUANTTYPE_Linear |
| QFactor | 4 |
| SuperCompression | 0 |
| SearchHeight | 16 |
| SearchWidth | 16 |
| Return | Description |
|---|---|
| S_OK | Successful. |
| < 0 | An error occurred. |
| VFW_E_WRONG_STATE | The filter is in the wrong state. |
Writes the values of all of the properties listed above to the system registry, so that the decoder will remember these settings the next time it is used.
This method is automatically called by the property page when you apply modified settings by clicking on the Apply or OK button.
| Return | Description |
|---|---|
| S_OK | Successful. |
| < 0 | An error occurred. |
* To increase compression performance, MPEG4 uses Temporal Compression. Temporal compression is achieved by only encoding the difference between successive frames instead of the frames themselves. This means that frames cannot be considered in isolation. A given picture is constructed from the prediction from a previous picture, and may be used to predict the next picture. An MPEG-4 video stream is comprised of I Frames, B Frames and P Frames:
I: (Intraframe) frames. These frames are also called Key Frames, they are encoded without reference to other frames and can be decoded independent of any other frame, and hence they allow random access. I frames are the largest frames in size and least compressed.
P: (Predictive) frames are encoded using the previous I / P frame as reference and can only be decoded correctly if the previous I / P is available. P frames are smaller and more compressed than I frames.
B: (Bi-directional) frames are encoded using information frame the previous I / P frame and the next I / P frame and can only be decoded correctly if the previous I / P frame and the next I / P frame are available. B frames are smaller than both P and I frames and they are the most compressed frames.
So to increase the compression of MPEG-4 more B and P frames should be used by adjusting PFrames and BFrames properties.