-
1.
公开(公告)号:US20240305801A1
公开(公告)日:2024-09-12
申请号:US18349076
申请日:2023-07-07
发明人: Zhenghao Chen , Roberto Gerson De Albuquerque Azevedo , Christopher Richard Schroers , Yang Zhang , Lucas Relic
IPC分类号: H04N19/42 , H04N19/172 , H04N19/91
CPC分类号: H04N19/42 , H04N19/172 , H04N19/91
摘要: In some embodiments, a system includes a first component to extract temporal features from a current frame being coded and a previous frame of a video. A second component uses a first transformer to fuse spatial features from the current frame with the temporal features to generate spatio-temporal features as first output. A third component uses a second transformer to perform entropy coding using the first output and at least a portion of the temporal features to generate a second output. A fourth component uses a third transformer to reconstruct the current frame based on the first output that is processed using the second output and the temporal features.
-
公开(公告)号:US20240303983A1
公开(公告)日:2024-09-12
申请号:US18599021
申请日:2024-03-07
发明人: Derek Edward BRADLEY , Prashanth CHANDRAN , Paulo Fabiano URNAU GOTARDO , Christopher Andreas OTTO , Gaspard ZOSS
CPC分类号: G06V10/993 , G06T17/205 , G06V10/82
摘要: One embodiment of the present invention sets forth a technique for evaluating three-dimensional (3D) reconstructions. The technique includes generating a 3D reconstruction of an object based on one or more mesh parameters. The technique also includes generating, based on the 3D reconstruction, a 3D rendering of the object. The technique further includes generating, using a machine learning model, a perceptual score associated with the 3D rendering and an input image of the object. The generated score represents how closely the 3D rendering matches the input image.
-
公开(公告)号:US12087024B2
公开(公告)日:2024-09-10
申请号:US16811219
申请日:2020-03-06
发明人: Abdelaziz Djelouah , Leonhard Markus Helminger , Scott Labrozzi , Yuanyi Xue , Erika Varis Doggett , Jared McPhillen , Christopher Richard Schroers
IPC分类号: H04N19/60 , G06T9/00 , H04N19/126 , H04N19/91
CPC分类号: G06T9/002 , H04N19/126 , H04N19/60 , H04N19/91
摘要: According to one implementation, an image compression system includes a computing platform having a hardware processor and a system memory storing a software code. The hardware processor executes the software code to receive an input image, transform the input image to a latent space representation of the input image, and quantize the latent space representation of the input image to produce multiple quantized latents. The hardware processor further executes the software code to encode the quantized latents using a probability density function of the latent space representation of the input image, to generate a bitstream, and convert the bitstream into an output image corresponding to the input image. The probability density function of the latent space representation of the input image is obtained based on a normalizing flow mapping of one of the input image or the latent space representation of the input image.
-
公开(公告)号:US12080929B2
公开(公告)日:2024-09-03
申请号:US17433355
申请日:2020-02-25
申请人: ETH Zürich
IPC分类号: H01M8/18 , C01B3/10 , C25B15/08 , H01M8/0656 , H01M8/22
CPC分类号: H01M8/186 , C01B3/10 , C25B15/081 , H01M8/0656 , H01M8/22
摘要: The present invention relates to energy storage systems and reactors useful in such systems. Inventive reactors comprise a reaction vessel defining an inner volume and a compensation element, whereby said inner volume is filled with a fixed bed that is free of cavities and that comprises particles of formula (I), FeOx (I), where 0≤x≤1.5; said compensation element is adapted to adjust said inner volume. The reactors are inherently explosion proof and thus suited for domestic use. The systems are useful for compensating long-term fluctuations observed in production of renewable energy.
-
公开(公告)号:US20240283957A1
公开(公告)日:2024-08-22
申请号:US18653776
申请日:2024-05-02
发明人: Abdelaziz Djelouah , Leonhard Markus Helminger , Roberto Gerson de Albuquerque Azevedo , Christopher Richard Schroers , Scott Labrozzi , Yuanyi Xue
IPC分类号: H04N19/42
CPC分类号: H04N19/42
摘要: A system includes a machine learning (ML) model-based video encoder configured to receive an uncompressed video sequence including multiple video frames, determine, from among the multiple video frames, a first video frame subset and a second video frame subset, encode the first video frame subset to produce a first compressed video frame subset, and identify a first decompression data for the first compressed video frame subset. The ML model-based video encoder is further configured to encode the second video frame subset to produce a second compressed video frame subset, and identify a second decompression data for the second compressed video frame subset. The first decompression data is specific to decoding the first compressed video frame subset but not the second compressed video frame subset, and the second decompression data is specific to decoding the second compressed video frame subset but not the first compressed video frame subset.
-
公开(公告)号:US20240282028A1
公开(公告)日:2024-08-22
申请号:US18172239
申请日:2023-02-21
发明人: Martin GUAY , Dhruv AGRAWAL , Dominik Tobias BORER , Jakob Joachim BUHMANN , Mattia Gustavo Bruno Paolo RYFFEL , Robert Walker SUMNER
CPC分类号: G06T13/00 , G06T7/248 , G06T7/74 , G06T2207/10016 , G06T2207/20081 , G06T2207/20084 , G06T2207/30196
摘要: One embodiment of the present invention sets forth a technique for training a neural motion controller. The technique includes determining a first set of features associated with a first control signal for a virtual character. The technique also includes matching the first set of features to a first sequence of motions included in a plurality of sequences of motions. The technique further includes training the neural motion controller based on one or more motions included in the first sequence of motions and the first control signal.
-
公开(公告)号:US20240249459A1
公开(公告)日:2024-07-25
申请号:US18421429
申请日:2024-01-24
发明人: Derek Edward BRADLEY , Prashanth CHANDRAN , Eftychios Dimitrios SIFAKIS , Barbara SOLENTHALER , Paulo Fabiano URNAU GOTARDO , Lingchen YANG , Gaspard ZOSS
IPC分类号: G06T13/40
CPC分类号: G06T13/40
摘要: One embodiment of the present invention sets forth a technique for retargeting a facial expression to a different facial identity. The technique includes generating, based on an input target facial identity, a facial identity code in an input identity latent space. The technique further includes converting a spatial input point from an input facial identity space of the input target facial identity to a canonical-space point in a canonical space. The technique still further includes generating one or more canonical simulator control values based on the facial identity code, an input source facial expression, and the canonical-space point. The technique still further includes generating a simulated active soft body based on one or more identity-specific control values, wherein each identity-specific control value corresponds to one or more of the canonical simulator control values and is in an output facial identity space associated with an output target facial identity.
-
公开(公告)号:US20240225700A1
公开(公告)日:2024-07-11
申请号:US18576883
申请日:2022-07-08
申请人: ETH ZÜRICH
CPC分类号: A61B17/68 , C22C1/02 , C22C23/00 , C22F1/06 , A61B2017/00004 , A61B2017/00526
摘要: A method of producing an alloy including magnesium and calcium, preferably an implantable medical device having magnesium and calcium, includes the steps of generating a billet including magnesium and calcium, and extruding the billet. The billet is extruded at least once at an extrusion temperature in the range of 250° C. to 450° C. and at a ram speed in the range of 0.01 mm/s to 1 mm/s and at an extrusion ratio in the range of 20 to 150 and preferably at an extrusion ratio in the range of 35 to 150.
-
公开(公告)号:US20240209357A1
公开(公告)日:2024-06-27
申请号:US18039578
申请日:2021-12-01
申请人: ETH ZÜRICH
发明人: Yaakov Benenson , Jlten Doshi
IPC分类号: C12N15/113 , C12N15/86
CPC分类号: C12N15/113 , C12N15/86 , C12N2310/14 , C12N2740/15043
摘要: The invention relates to the fields of biomolecular computing and/or synthetic biology. In particular, the invention relates to a DNA construct that comprises a first promoter, at least one further promoter and an output sequence, wherein each of said promoters comprises a transcription start site and/or is suitable for initiating transcription, wherein the initiation of transcription from the first promoter is enabled by a first transcriptional regulatory state and the initiation of transcription from each of said further promoter(s) is enabled by a respective further transcriptional regulatory state, and wherein said DNA construct yields an effective amount of an output RNA in a eukaryotic cell, when said first transcriptional regulatory state and/or any of the respective further transcriptional regulatory states is present in said cell, wherein said output RNA comprises a sequence corresponding to said output sequence. Furthermore, the invention relates to medical and/or diagnostic uses of the inventive DNA construct of the invention, e.g., for detecting, killing and/or manipulating different types of eukaryotic target cells in a subject and/or in a tissue sample.
-
公开(公告)号:US20240171138A1
公开(公告)日:2024-05-23
申请号:US18263140
申请日:2022-02-03
申请人: ETH Zürich
CPC分类号: H03F3/2178 , H02M3/338
摘要: A DC-DC converter topology based on electromagnetically coupled class-D LC oscillator is proposed. An electrical converter comprises at least two oscillators (1,2), each of the at least two oscillators being designed to have an oscillating current and an oscillating voltage. Coupling elements (16, 17, 18, 19) arranged to couple the oscillating currents of the at least two oscillators and/or the oscillating voltages of the at least two oscillator. The at least two oscillators are connected in a series connection, adding their oscillating voltages, and/or in a parallel connection, adding their oscillating currents. The topology can be fully integrated, that is, it can be realized as an integrated circuit without external components, in particular without external passive components, such as capacitors and/or inductors.
-
-
-
-
-
-
-
-
-