Abstract: The present disclosure provides a method for visual question answering, which relates to fields of computer vision and natural language processing. The method includes: acquiring an input image and an input question; detecting visual information and position information of each of at least one text region in the input image; determining semantic information and attribute information of each of the at least one text region based on the visual information and the position information; determining a global feature of the input image based on the visual information, the position information, the semantic information, and the attribute information; determining a question feature based on the input question; and generating a predicted answer for the input image and the input question based on the global feature and the question feature. The present disclosure further provides a device for visual question answering, a computer device and a medium.

Abstract: In a method for processing a document image, a document image to be processed is acquired. Text nodes of multiple granularities, visual nodes of multiple granularities, respective node information of the text nodes, and respective node information of the visual nodes in the document image are obtained. A multi-granularity and multi-modality document graph is construct based on the text nodes of multiple granularities, the visual nodes of multiple granularities, the respective node information of the text nodes and the respective node information of the visual nodes. Multi-granularity semantic feature information of the document image is determined based on the multi-granularity and multi-modality document graph, the respective node information of the text nodes and the respective node information of the visual nodes.

Abstract: A neural network training method and a document image understanding method is provided. The neural network training method includes: acquiring text comprehensive features of a plurality of first texts in an original image; replacing at least one original region in the original image to obtain a sample image including a plurality of first regions and a ground truth label for indicating whether each first region is a replaced region; acquiring image comprehensive features of the plurality of first regions; inputting the text comprehensive features of the plurality of first texts and the image comprehensive features of the plurality of first regions into a neural network model together to obtain text representation features of the plurality of first texts; determining a predicted label based on the text representation features of the plurality of first texts; and training the neural network model based on the ground truth label and the predicted label.

Abstract: The present disclosure provides a training method and apparatus for a document processing model, a device, a storage medium and a program, which relate to the field of artificial intelligence, and in particular, to technologies such as deep learning, natural language processing and text recognition. The specific implementation is: acquiring a first sample document; determining element features of a plurality of document elements in the first sample document and positions corresponding to M position types of each document element according to the first sample document; where the document element corresponds to a character or a document area in the first sample document; and performing training on a basic model according to the element features of the plurality of document elements and the positions corresponding to the M position types of each document element to obtain the document processing model.

Abstract: The present disclosure provides a method for visual question answering, which relates to fields of computer vision and natural language processing. The method includes: acquiring an input image and an input question; detecting visual information and position information of each of at least one text region in the input image; determining semantic information and attribute information of each of the at least one text region based on the visual information and the position information; determining a global feature of the input image based on the visual information, the position information, the semantic information, and the attribute information; determining a question feature based on the input question; and generating a predicted answer for the input image and the input question based on the global feature and the question feature. The present disclosure further provides a device for visual question answering, a computer device and a medium.

Abstract: Disclosed is a double-sided organic light-emitting diode lighting panel. A semitransparent silver film layer is provided between a first light-emitting layer that emits blue light and a second light-emitting layer that emits yellow light. Double-sided light emission of the OLED lighting panel is realized by taking advantage of semi-transparence and reflection properties of the silver film layer. Such a structure is relatively simple, and manufacture costs and manufacture difficulties are reduced. The OLED lighting panel can be used for interior lighting or can be used in places having special needs, and thus application of a double-sided OLED lighting device is widened.

Abstract: A thin-film package structure and an OLED component are disclosed. The present disclosure aims to solve the technical problem of a low light-emitting efficiency of an OLED display device and a poor package effect in the prior art. The thin-film package structure includes at least two organic material film layers, and at least two inorganic material film layers. An organic material film layer and an inorganic material film layer are arranged in an alternate manner. The inorganic material film layer includes a magnetic material film layer and a non-magnetic material film layer. The OLED component packaged by the thin-film package structure includes a substrate, and an OLED component that is formed on the substrate. A magnetic field generated by the magnetic material in the thin-film layers can act on the OLED, and thus a light-emitting efficiency of the OLED display device can be improved.

Abstract: Disclosed is a top-emitting WOLED display device. An optical turning layer with a uniform thickness is added between anode layers of three sub-pixels R, G and B and a reflective metal layer of the display device. Because of arrangement of the optical turning layer, a distance between a WOLED layer and a reflective metal layer is increased. Thus, a total cavity length of a white OLED is increased, and a microcavity effect of a top-emitting WOLED is overcome. The optical turning layer is suitable for manufacture of AMOLED display devices with high-resolution and color gamut.

Abstract: Disclosed is a double-sided organic light-emitting diode lighting panel. A semitransparent silver film layer is provided between a first light-emitting layer that emits blue light and a second light-emitting layer that emits yellow light. Double-sided light emission of the OLED lighting panel is realized by taking advantage of semi-transparence and reflection properties of the silver film layer. Such a structure is relatively simple, and manufacture costs and manufacture difficulties are reduced. The OLED lighting panel can be used for interior lighting or can be used in places having special needs, and thus application of a double-sided OLED lighting device is widened.

Abstract: Disclosed is a top-emitting WOLED display device. An optical turning layer with a uniform thickness is added between anode layers of three sub-pixels R, G and B and a reflective metal layer of the display device. Because of arrangement of the optical turning layer, a distance between a WOLED layer and a reflective metal layer is increased. Thus, a total cavity length of a white OLED is increased, and a microcavity effect of a top-emitting WOLED is overcome. The optical turning layer is suitable for manufacture of AMOLED display devices with high-resolution and color gamut.

Abstract: A thin-film package structure and an OLED component are disclosed. The present disclosure aims to solve the technical problem of a low light-emitting efficiency of an OLED display device and a poor package effect in the prior art. The thin-film package structure includes at least two organic material film layers, and at least two inorganic material film layers. An organic material film layer and an inorganic material film layer are arranged in an alternate manner. The inorganic material film layer includes a magnetic material film layer and a non-magnetic material film layer. The OLED component packaged by the thin-film package structure includes a substrate, and an OLED component that is formed on the substrate. A magnetic field generated by the magnetic material in the thin-film layers can act on the OLED, and thus a light-emitting efficiency of the OLED display device can be improved.

Abstract: The present application discloses an OLED display device and its manufacturing method thereof, the OLED display device includes a substrate; an anode formed on the substrate; a hole injection layer formed on the anode; a hole transport layer formed on the hole injection layer; a light emitting layer formed on the hole transport layer; a hole blocking layer formed on the light emitting layer; an electron transporting layer formed on the hole blocking layer; an electron injection layer formed on the electron transporting layer; a cathode formed on the electron injection layer; a cover plate disposed opposite to and cover the substrate; and a sealing adhesive material formed between the edge of the substrate and the cover plate; wherein the material of the electron transporting layer is a mixture of an organic electron transporting material and an organometal halide perovskite material.