Abstract: Methods and systems for budgeted and simplified training of deep neural networks (DNNs) are disclosed. In one example, a trainer is to train a DNN using a plurality of training sub-images derived from a down-sampled training image. A tester is to test the trained DNN using a plurality of testing sub-images derived from a down-sampled testing image. In another example, in a recurrent deep Q-network (RDQN) having a local attention mechanism located between a convolutional neural network (CNN) and a long-short time memory (LSTM), a plurality of feature maps are generated by the CNN from an input image. Hard-attention is applied by the local attention mechanism to the generated plurality of feature maps by selecting a subset of the generated feature maps. Soft attention is applied by the local attention mechanism to the selected subset of generated feature maps by providing weights to the selected subset of generated feature maps in obtaining weighted feature maps.

Abstract: The present disclosure relates to compositions and methods for enhancing infiltration of lymphocytes into tumor tissue, enhancing anti-tumor lymphocyte activities in tumor microenvionment (TME), inhibiting regulatory lymphocyte (e.g., B and T cells) activities in TME, and/or long term benefit of cell therapies. For example, in a method of in vivo cell expansion, the method comprises administering an effective amount of cells comprising an antigen binding molecule to a subject; and administering an effective amount of presenting cells expressing a solid tumor antigen that the binding molecule binds.

Abstract: A method and an apparatus for calling a terminal to join a conference, related to the field of communications technologies, resolving a problem that a process in which a calling terminal calls a called terminal to join the conference or a called terminal requests joining the conference is complicated. The method includes the following steps: receiving first call request information, where the first call request information carries identification information of the called terminal; and calling the called terminal according to the identification information of the called terminal, so that the called terminal receives a message which is published by another terminal in a current conference and that the called terminal publishes a message to another terminal in the current conference. The method may be applied to calling a terminal to join a conference.

Abstract: An audio mixing method, apparatus and system, which can ensure sound quality after audio mixing and reduce consumption of computing resources. The method includes: receiving an audio stream of each site, and analyzing the audio stream of each site to obtain a sound characteristic value of a sound source object; selecting, according to a descending sequence of sound characteristic values of sound source objects, a predetermined number of sound source objects from the sound source objects to serve as main sound source objects; determining, according to a relationship between a target site and the sites where the main sound source objects are located, audio streams that require audio mixing for the target site; and performing audio mixing on the audio streams that require audio mixing for the target site and sending the audio streams after the audio mixing to the target site.

Abstract: A method and an apparatus for playing audio of an attendant at a remote end and a remote video conference system are provided. The method includes: receiving audio of an attendant at a remote site; and by means of two or more loudspeakers mounted at a top and a bottom of a remote image presentation device at a local site, simulating an audio transmission path of the audio of the attendant at the remote site between the two or more loudspeakers and an attendant at the local site through a predetermined algorithm, where the simulating means that the audio transmission path is simulated between an image of the head of the attendant at the remote site displayed in the remote image presentation device, and the head of the attendant at the local site.

Abstract: A method, an apparatus, and a system for controlling multipicture display. The method includes receiving a sliding track of a user operation, and performing an operation on a multipicture division line according to the sliding track of the user operation; and controlling picture display on a target screen according to a multipicture display mode formed using the multipicture division line that is obtained after the operation. The method, apparatus, and system for controlling multipicture display according to the embodiments of the present invention can implement visual and flexible setting of a multipicture display mode.

Abstract: The present invention discloses a signal processing method, system and apparatus for 3-dimensional (3D) audio conferencing. The implementation is: a server obtains at least one audio stream relative to one terminal; the server allocates identifiers for the obtained at least one audio stream relative to the terminal; and the server combines the obtained at least one audio stream and the identifiers of the at least one audio stream and sends the combination to the terminal. With the technical solution of the present invention, the issue of excessive transmission channels required in the prior art is resolved and the terminal is capable of determining the sound image positions of other terminals freely.

Abstract: A method and an apparatus for calling a terminal to join a conference, related to the field of communications technologies, resolving a problem that a process in which a calling terminal calls a called terminal to join the conference or a called terminal requests joining the conference is complicated. The method includes the following steps: receiving first call request information, where the first call request information carries identification information of the called terminal; and calling the called terminal according to the identification information of the called terminal, so that the called terminal receives a message which is published by another terminal in a current conference and that the called terminal publishes a message to another terminal in the current conference. The method may be applied to calling a terminal to join a conference.

Abstract: A conference terminal, a conference server, a conference system, and a data processing method are provided. The conference terminal includes: a mark obtaining module, configured to obtain a mark data signal; an image obtaining module, configured to obtain an image data signal; a superimposing module, configured to superimpose the mark data signal onto the image data signal to generate a hybrid image data signal; and an outputting module, configured to output the hybrid image data signal. The present invention improves the interactivity between videoconference users and the discussion efficiency.

Abstract: An audio mixing method, apparatus and system, which can ensure sound quality after audio mixing and reduce consumption of computing resources. The method includes: receiving an audio stream of each site, and analyzing the audio stream of each site to obtain a sound characteristic value of a sound source object; selecting, according to a descending sequence of sound characteristic values of sound source objects, a predetermined number of sound source objects from the sound source objects to serve as main sound source objects; determining, according to a relationship between a target site and the sites where the main sound source objects are located, audio streams that require audio mixing for the target site; and performing audio mixing on the audio streams that require audio mixing for the target site and sending the audio streams after the audio mixing to the target site.

Abstract: A method, an apparatus, and a system for implementing audio mixing are provided. The method for implementing audio mixing includes: receiving an audio signal sent by each sending conference site, where the audio signals include an audio channel-based audio signal and an audio object-based audio signal; selecting an audio signal for each receiving conference site from the received audio signals; processing the selected audio signals according to types of receiving conference sites, where the receiving conference sites include an audio channel-based receiving conference site and an audio object-based receiving conference site; and sending the processed audio signal to each receiving conference site respectively according to the type of the receiving conference site. The present invention is applicable to a multipoint conference where audio mixing is performed on an audio channel-based audio signal and an audio object-based audio signal.

Abstract: The embodiments of the present invention disclose a method, device, and system for video communication, and relate to the field of video conference technologies, so as to implement automatic switching of video images during a video conference. The method includes: obtaining video image signals and voice information of a first site; determining video image signals including a video object according to the video image signals and voice information of the first site; and sending the video image signals including the video object to a second site. The method, device, and system provided in the embodiments of the present invention implement automatic switching of video image signals during the video conference, thus avoiding switching errors caused by human factors and improving the efficiency of the conference.

Abstract: A method for generating and playing audio signals and a system for processing audio signals are disclosed. The method for generating audio signals includes: generating distance information about an audio signal corresponding to a view point position, according to obtained auxiliary video and direction information about the audio signal, where the auxiliary video is a disparity map or a depth map; encoding the direction information and distance information about the audio signal, and sending the encoded information. The apparatus for generating audio signals includes an audio signal distance information obtaining module and an audio signal encoding module. With the present invention, the position information, including direction information and distance information, about the audio signal may be obtained accurately in combination with a three-dimensional video signal and a three-dimensional audio signal, without increasing the size of a microphone array, and the audio signal is sent and played.

Abstract: A method, device and system to implement hiding the loss packet are provided. The provided method, device and system recover the lost frame according to the data before and after the lost frame and enhances the correlation of the recovered lost frame data and the data after the lost frame. A method and device for estimating pitch period are also provided which select a pitch period from the initial pitch period and the pitch periods corresponding to the frequencies which are one or more times higher than the frequencies corresponding to the initial pitch period as the final estimated pitch period, may improve frequency multiplication when estimating the pitch period; in addition, by tuning of the pitch period by matching the waves, the error of estimating pitch period may be reduced and the quality of the audio data may be improved.

Abstract: A method for concealing lost frame includes: using history signals before the lost frame that corresponds to a lost MDCT coefficient to generate a first synthesized signal when it is detected that the MDCT coefficient is lost; performing fast IMDCT for the first synthesized signal to obtain an IMDCT coefficient corresponding to a lost MDCT coefficient; and using the IMDCT coefficient corresponding to the lost MDCT coefficient and an IMDCT coefficient adjacent to the IMDCT coefficient corresponding to the lost MDCT coefficient to perform TDAC and obtain signals corresponding to the lost frame. An apparatus for concealing lost frame is also disclosed herein. The method and the apparatus for concealing lost frames in the embodiments of the present invention make full use of the received partial signals to recover high-quality voice signals and improve the QoS.

Abstract: A device for lost frame concealment comprises: a lost frame detector for detecting whether a voice frame is lost, a decoding module for decoding the current voice frame, a low band delay module for delaying the low band signal, a low band signal recovering module for recovering the lost low band signal, a high band lost frame concealment module for processing the lost frame concealment for the high band signal, and a QMF synthesis filter for synthetically filtering the low band signal and the high band signal. The invention makes full use of the delay of the coding/decoding device itself, enhances the effect of lost frame concealment for the low band signal and the high band signal, and introduces no nearby delay during the process of lost frame concealment.

Abstract: The present invention discloses a method for obtaining an attenuation factor. The method is adapted to process the synthesized signal in packet loss concealment, and includes: obtaining a change trend of a signal; obtaining an attenuation factor, according to the change trend of the signal. The present invention also discloses an apparatus for obtaining an attenuation factor. A self-adaptive attenuation factor is adjusted dynamically by using the latest change trend of a history signal by using the present invention. The smooth transition from the history data to the data last received is realized so that the attenuation speed is kept consistent between the compensated signal and the original signal as much as possible for adapting to the characteristic of various human voices.

Abstract: A method, an apparatus, and a system for implementing audio mixing are provided. The method for implementing audio mixing includes: receiving an audio signal sent by each sending conference site, where the audio signals include an audio channel-based audio signal and an audio object-based audio signal; selecting an audio signal for each receiving conference site from the received audio signals; processing the selected audio signals according to types of receiving conference sites, where the receiving conference sites include an audio channel-based receiving conference site and an audio object-based receiving conference site; and sending the processed audio signal to each receiving conference site respectively according to the type of the receiving conference site. The present invention is applicable to a multipoint conference where audio mixing is performed on an audio channel-based audio signal and an audio object-based audio signal.

Abstract: The present invention discloses a method for performing a frame erasure concealment to a higher-band signal, including: calculating a periodic intensity of a higher-band signal with respect to a lower-band signal; judging whether the periodic intensity of the higher-band signal is higher than or equal to a preconfigured threshold; if the periodic intensity of the higher-band signal is higher than or equal to the preconfigured threshold, using a pitch period repetition method to perform the frame erasure concealment to the higher-band signal of a current lost frame; and if the periodic intensity of the higher-band signal is lower than the preconfigured threshold, using a previous frame data repetition method to perform the frame erasure concealment to the higher-band signal of the current lost frame. The present invention further discloses a device for performing a frame erasure concealment to a higher-band signal and a speech decoder. The problem that the quality of the voice signal is lowered is avoided.

Abstract: The embodiments of the present invention disclose a method, device, and system for video communication, and relate to the field of video conference technologies, so as to implement automatic switching of video images during a video conference. The method includes: obtaining video image signals and voice information of a first site; determining video image signals including a video object according to the video image signals and voice information of the first site; and sending the video image signals including the video object to a second site. The method, device, and system provided in the embodiments of the present invention implement automatic switching of video image signals during the video conference, thus avoiding switching errors caused by human factors and improving the efficiency of the conference.