o ϯi,@sddlZddlZddlmZmZmZmZmZddlZddl Z ddl Z ddl m Z ddlmZmZmZeeZGddde Z ddejjjd ed ed ejjjd ed edeeeffddZdS)N)BinaryIODictListOptionalTuple) EncodedVideo) pts_to_secs secs_to_pts thwc_to_cthwc @seZdZdZ  ddedeededdfdd Ze deefd d Z e de fd d Z de de de eeejffddZddZdejfde de de fddZdS)EncodedVideoPyAVz EncodedVideoPyAV is an abstraction for accessing clips from an encoded video using PyAV as the decoding backend. It supports selective decoding when header information is available. NTfile video_name decode_audioreturnc Cs||_||_zt||_Wnty$}z td|d|d}~ww|jdus3t|jjj dkr:td||jjj d}|j |_ |j |_ |j durQd|_ |j}d}d|_|jr|jjj|_|jr|jjjdj |_|jjjdj |_|jdurd|_|jjjdj}d\|_|_|_|dur|durd|_|\|_|_|jdurtd |jd d }|jdur|jd d }|dur|durtt||j |j t||j|j|_dS|durt||j |j |_dS|durt||j|j|_dSdS) z Args: file (BinaryIO): a file-like object (e.g. io.BytesIO or io.StringIO) that contains the encoded video. zFailed to open video . NrzVideo stream not found )NNTFzUnable to decode video streamr) _video_name _decode_audioavopen _container Exception RuntimeErrorlenstreamsvideo time_base_video_time_base start_time_video_start_ptsduration _has_audioaudio_audio_time_base_audio_start_pts_video_audio_selective_decoding_pyav_decode_videomaxr _duration)selfr rre video_streamvideo_durationaudio_durationr2X/home/ubuntu/.local/lib/python3.10/site-packages/pytorchvideo/data/encoded_video_pyav.py__init__sj             zEncodedVideoPyAV.__init__cC|jS)zQ Returns: name: the name of the stored video if set. )rr-r2r2r3namefzEncodedVideoPyAV.namecCr5)zZ Returns: duration: the video's duration/end-time in seconds. )r,r6r2r2r3r"nr8zEncodedVideoPyAV.duration start_secend_secc s2|jr |||\|_|_d}|jdur3t||j|jddt||j|jddfdd|jD}d}|jri|jdurit||j|j ddt||j|j ddfdd|jD}t j |dd}| t j }|dusst|dkrtd |d |d |jd d}|durtt | t j }||d S)a Retrieves frames from the encoded video at the specified start and end times in seconds (the video always starts at 0 seconds). Returned frames will be in [start_sec, end_sec). Note that 1) if you want to avoid float precision issue and need accurate frames, please use Fraction for start_sec and end_sec. 2) As end_sec is exclusive, so you may need to use `get_clip(start_sec, duration + EPS)` to get the last frame. Args: start_sec (float): the clip start time in seconds end_sec (float): the clip end time in seconds Returns: clip_data: A dictionary mapping the entries at "video" and "audio" to a tensors. "video": A tensor of the clip's RGB frames with shape: (channel, time, height, width). The frames are of type torch.float32 and in the range [0 - 255]. "audio": A tensor of the clip's audio samples with shape: (samples). The samples are of type torch.float32 and in the range [0 - 255]. Returns None if no video or audio found within time range. Nceil round_modec$g|]\}}|kr|kr|qSr2r2.0fpts) video_end_ptsvideo_start_ptsr2r3 z-EncodedVideoPyAV.get_clip..cr>r2r2r?) audio_end_ptsaudio_start_ptsr2r3rErFraxiszNo video found within z and z- seconds. Video starts at time 0 and ends at .)rr$)r)r*r'r(r rr!r#r%r&torchcattofloat32rloggerdebugr"r stack)r-r9r: video_frames audio_samplesr2)rGrHrCrDr3get_clipvsd    zEncodedVideoPyAV.get_clipcCs|jdur |jdSdS)z6 Closes the internal video container. N)rcloser6r2r2r3rVs zEncodedVideoPyAV.closer start_secsend_secsc Cs,d}d}zpt|jt||j|jddt||j|jdd|jjjdddi\}}t|dkr4dd|D}|jrkt|jt||j |j ddt||j |j dd|jjj dddi\}}t|dkrpd d|D}W||fSW||fSW||fSt y}zt d |jd |WYd}~||fSd}~ww) z Selectively decodes a video between start_pts and end_pts in time units of the self._video's timebase. Nr;r<rrcSs$g|]}t||jfqSr2)rL from_numpyto_rgb to_ndarrayrBr@framer2r2r3rEsz7EncodedVideoPyAV._pyav_decode_video..r$cSs*g|]}ttj|dd|jfqS)rrI)rLrYnpmeanr[rBr\r2r2r3rE s zFailed to decode video: r)_pyav_decode_streamrr rr!rrrr#r%r&r$rrPrQr) r-rWrX video_and_pts audio_and_ptspyav_video_frames_pyav_audio_framesr.r2r2r3r*sp      "z#EncodedVideoPyAV._pyav_decode_video)NT)__name__ __module__ __qualname____doc__rrstrboolr4propertyr7floatr"rrLTensorrUrVmathinfr*r2r2r2r3r sB  M [r container start_ptsend_ptsstream stream_name buffer_sizerc sd}t||d}|jt|dd|did}|jd i|D]} t|| j}| j|kr8| j|kr8| | j<q | j|kr?nq fddtD} | |fS) aF Decode the video with PyAV decoder. Args: container (container): PyAV container. start_pts (int): the starting Presentation TimeStamp to fetch the video frames. end_pts (int): the ending Presentation TimeStamp of the decoded frames. stream (stream): PyAV stream. stream_name (dict): a dictionary of streams. For example, {"video": 0} means video stream at stream index 0. Returns: result (list): list of decoded frames. max_pts (int): max Presentation TimeStamp of the video sequence. irFT) any_framebackwardrtcsg|]}|qSr2r2)r@rBframesr2r3rE@sz'_pyav_decode_stream..Nr2)r+seekintdecoderBsorted) rqrrrsrtrurvmargin seek_offsetmax_ptsr]resultr2ryr3r`s   r`)r) loggingrotypingrrrrrrnumpyr^rLpytorchvideo.data.encoded_videorutilsr r r getLoggerrfrPr rqinputInputContainerr|rrt VideoStreamdictrmr`r2r2r2r3s8