Abstract: The present invention relates to the field of artificial intelligence, and in particular, to a camera pose estimation method and system, electronic equipment and a readable medium. The camera pose estimation method includes: acquiring an initial matching set between a first image and a second image, where the first image and the second image are images from different angles for a same scene; performing a mismatch removal operation on the initial matching set based on an optimization network to obtain an optimized matching set, where the optimization network is constructed based on a multi-stage geometric semantic attention network; and acquiring a camera pose result based on the optimized matching set. By removing mismatches, feature matching results between the first image and the second image are more accurate, and thus a more accurate result can be obtained when performing camera pose estimation.

Abstract: Provided are viral vector and recombinant Simian adenovirus. The viral vector carries at least one recombinant genome preparing from the Simian adenovirus genome by replacing E4orf6 gene, knocking out E3 gene, inserting EGFP gene at the loci of E3, or knocking out E1B55K gene. Recombinant Simian adenovirus could be preparing by transfecting the viral vector into host cells for packaging.

Abstract: This invention discloses a multimodal data heterogeneous Transformer-based asset recognition method, system, and device, the method including: collecting various-modal information of an asset, including text information and image information; building an ALBERT model, a ViT model, and a CLIP model; by the ALBERT model, extracting a text information feature; by the ViT model, extracting an image information feature; by the CLIP model, extracting image-text matching information feature; by different channels, applying asset type recognition to information in different modalities; outputting classification information from the different channels; by the CLIP model, generating asset void information; and discriminatively fusing the classification information from the different channels with the matching degree between the image information and the text information obtained by the CLIP model, and outputting final asset class information.

Abstract: A system for enabling escrow-free, secure, and convenient atomic asset swaps between blockchains are provided. In detail, when any group of users reach an agreement on the content of cross-chain asset swap transactions, each participant creates an individual signature on the content of each transaction, employs a reversible signature aggregate function to produce an aggregate signature from all his/her individual signatures, and optionally generates a proof to prove the relationship between the aggregated signature and the transaction contents. Sequentially, all the participants exchange these aggregated signatures and optional proofs. Once a participant publishes one of his/her individual signature, the other parties can use both signatures—one from the sender and the other from the recipient—to construct a complete transaction. Once a transaction is broadcast, its corresponding transactions are constructed, and broadcast.

Abstract: A fault-tolerant asset transferring method is provided to prevent financial losses caused by human errors, specifically unintentional mistakes made by the payer. The system introduces a new definition for blockchain transaction parameters, allowing for multi-input transactions. Each payment transaction includes the payee's signature for confirmation and the payer's signature. In addition, a system is designed to handle multi-input asset transfer transactions which incorporate the payer's confirmation. Upon reaching an agreement on the transaction content, each participant generates an independent off-chain signature on the transaction's content. In the event that participants generate a signature on the erroneous content of a transaction, the transaction will be accepted in the blockchain with very low probability due to transaction validation.

Abstract: A method for detecting a state of an energy storage device (e.g. lithium-ion battery) is provided, which includes acquiring at least one of a change signal of a temperature or a change signal of a pressure inside the energy storage device, and determining the state of the energy storage device based thereon. The method may further include triggering a warning when the state of the energy storage device meets a preset condition, such as when the energy storage device is determined to be in an irreversible state, an internal short circuit state, a safety valve opening state, or a thermal runaway state. A detection device implementing the detection method is also provided, which comprises a sensing module and an analyzing module. The sensing module is arranged at an interior position of the energy storage device, and can include an optical sensor and/or an electrical sensor.

Abstract: The present invention relates to the technical field of active substance development, and especially relates to the use of Fructus lycii glucopeptide. In the present invention, the protection effect of the Fructus lycii glucopeptide on Parkinson-like dyskinesia is studied by means of using a Parkinson's disease animal model as a study object. The result shows that the Fructus lycii glucopeptide can improve the Parkinson-like dyskinesia of the model animal.

Abstract: A multi-source image correspondence method and system based on heterogeneous model fitting is provided, the method includes the following steps: constructing a multi-orientation phase consistency model, fusing phase consistency, image amplitude, and orientation detection feature points, constructing logarithmic polar coordinate descriptors with variable-size bins using sub-region grids and orientation histograms, effectively estimating model parameters through heterogeneous model fitting, accumulating matching pairs from different heterogeneous models that meet a preset joint position offset transformation error, outputting a final matching pair, and completing multi-source image correspondence.

Abstract: Disclosed in the present invention is the use of Crassifolin A in promoting mitophagy and treating neurodegenerative diseases. According to the present invention, experiments show that the diterpene compound Crassifolin A can significantly increase mitophagy levels, promote the removal of damaged mitochondria, maintain homeostasis in cells, and has potential effects of treating neurodegenerative diseases and other related diseases; moreover, Crassifolin A is a small-molecule compound that can easily penetrate through the blood-brain barrier to reach the brain so as to play a role, can be extracted from a Thickleaf croton root medicinal material, has the advantages of abundant medicinal sources, easy availability, low drug toxicity and few side effects, and has a good application prospect in the treatment of neurodegenerative diseases.

Abstract: The embodiment of the invention discloses a diversified recommendation method for news based on graph neural network and its device, which comprises the followings: word segmentation processing is performed on the target news text to obtain a word segmentation set, and vectorization processing is performed on the word segmentation set, so that each word segmentation in the word segmentation set has a word segmentation embedding vector; vectorization processing is performed on a target user group, so that each user in the target user group has a user embedding vector; the word segmentation embedding vector and the user embedding vector are input into the attention mechanism algorithm to obtain the news embedding vector; a graph neural network is constructed according to the user embedding vector and the news embedding vector, and the recommendation model is trained to obtain diversified recommendation results.

Abstract: A specific target-oriented social media tweet sentiment analysis method includes: preprocessing social media tweet data to obtain a target text and a specific target; passing the target text and the specific target through an embedding layer to obtain target text word vectors and obtain a specific target word vector, respectively; passing the target text word vectors through a self-attention structure to obtain a self-attention result; combining the self-attention result with the specific target word vector and passing through a cross-attention structure to obtain cross-attention results; concatenating the cross-attention results to obtain an attention representation matrix; and passing the attention representation matrix through a pooling layer, a fully connected layer and a softmax layer to obtain a sentiment tendency of the specific target.

Abstract: The present disclosure provides a class of Pyridopyrimidine compounds having a structure shown in Formula (I) or their pharmaceutically acceptable salts, or stereoisomers or prodrug molecules and applications thereof. The compounds in the present disclosure can efficiently and selectively degrade AKT3 protein in cells without affecting AKT1/2, thereby significantly inhibiting tumor cell proliferation mediated by high expression of AKT3 protein. It can be used to prepare therapeutic drugs for cancer and other diseases related to abnormal expression of AKT3 protein.

Abstract: Disclosed are a neural network model update method and device, and a computer storage medium. The method includes: randomly generating a preset number of sub-neural network models as nodes in a blockchain; using a ring signature to share a local data set in the blockchain, and uniformly dividing a data set in the blockchain to generate a training set and a test set; training each node through the training set to generate a trained model, packaging the trained model as a model transaction and sharing the model transaction in the blockchain; using the test set selected by voting to test the model transaction and generating a test result; when the test result is greater than a benchmark evaluation, taking the sub-neural network model corresponding to the test result as a valid vote; and voting a previous block corresponding to the valid vote, selecting a consistent block, and updating all nodes.

Abstract: An in-situ and in-operando internal state detecting system for an energy storage device comprises a first acoustic signal-emitting acoustic signal emission module, an interior-residing optical sensing module, and a first input optical signal-providing signal demodulation module. The optical sensing module receives a second acoustic signal when the first acoustic signal transmits inside the energy storage device, and outputs a first optical signal, which is demodulated and analyzed by the signal demodulation module to obtain acoustic feature information. Optionally, the signal demodulation module also provides a second input optical signal, and the optical sensing module outputs a second optical signal, which is analyzed by the signal demodulation modulation module to obtain optical feature information.

Abstract: The present disclosure provides a chitosan-modified opto-hydrodynamic micromotor and a preparation method and use thereof, and belongs to the technical field of micromotors. The chitosan-modified opto-hydrodynamic micromotor provided by the present disclosure is composed of Phaeodactylum tricornutum Bohlin and a chitosan solution. The chitosan-modified opto-hydrodynamic micromotor provided by the present disclosure has high biocompatibility, can non-invasively remove biological threats in a microenvironment containing cells with a sterilization rate reaching about 98%, and can achieve high-efficiency sterilization without affecting cell viability.

Abstract: A respirator for measuring a respiratory rate of a biofilm includes a body, a carbon dioxide absorption tube and a perforated partition disposed in an interior of the body. The perforated partition divides the interior of the body into an oxygen mass-transfer stir zone including a stirring device and a biofilm reaction zone for placing MBBR fillers. The carbon dioxide absorption tube includes an air vent and contains a solution capable of absorbing CO2. The perforated partition has a first hole corresponding to a middle portion and a lower portion of the perforated partition and acting as a liquid-exchange channel, and a second hole corresponding to an upper portion of the perforated partition and acting as a gas-exchange channel, and the oxygen mass-transfer stir zone is communicated with the biofilm reaction zone through the liquid-exchange channel and the gas-exchange channel.

Abstract: Disclosed is a method for constructing an auditable and privacy-preserving collaborative deep learning platform based on a blockchain-empowered incentive mechanism, which allows trainers of multiple similar models to cooperate for training deep learning models while protecting confidentiality and auditing correctness of shared parameters. The invention has the following technical effects. Firstly, the encryption method used by model trainers protects the confidentiality of sharing parameters; furthermore, the updated parameters are decrypted through the cooperation of all participants, which reduces the possible disclosure of parameters. Secondly, the encrypted parameters are stored in the blockchain, and are only available to participants and authorized miners who are responsible to update parameters.

Abstract: The present disclosure provides a class of 2-aminopyrimidine compounds with a structure shown in Formula (I) or their pharmaceutically acceptable salts, isotope derivatives, solvates, or their stereoisomers, geometric isomers, tautomers, or prodrug molecules or metabolites and pharmaceutical compositions and application thereof. The compounds of the present disclosure can efficiently and selectively inhibit the kinase activity of Janus Kinase 3 (JAK3), and have strong signal inhibition and cell proliferation inhibition effects on various blood tumor cells (especially human acute myeloid leukemia cell U937 cells) and solid tumor cells. They can be used to prepare anti-tumor drugs and drugs for preventing and treating inflammatory diseases.

Abstract: Disclosed is a method for constructing an emergency reporting system of vehicle networking based on multi-signatures, which comprises the following steps: initializing an emergency reporting system to obtain a public-private key pair and pre-calculation parameters of a vehicle set; obtaining authentication information of the emergency based on the public-private key pair and the pre-calculation parameters; integrating the authentication information to obtain multi-signatures of the emergency; and completing the system construction by regulating a traffic state based on the multi-signatures. The method realizes the efficient signature generation function of the vehicle with limited computing resources, reduces the communication pressure and computing pressure of roadside units, and may effectively prevent the common rogue key attacks in other schemes.

Abstract: A dual-pulse excitation method for ultra-fast, super-resolution all-optical magnetic recording includes the steps of: providing a first excitation pulse and a second modulation pulse; and focusing the first excitation pulse and the second modulation pulse, and then radiating the two pulses in sequence to a magneto-optical recording medium, so that an area of the magneto-optical recording medium irradiated undergoes opto-magnetic reversal. By controlling the time delay, spatial overlapping area, and energy density ratio between the dual femtosecond laser pulses, it can induce a second reversal of the magnetization field in the spatial overlapping area of the two pulses on the magneto-optical material that can achieve single-pulse opto-magnetic reversal to obtain all-optical magnetic recording beyond the diffraction limit. This process takes place within several hundred picoseconds, thus providing an effective technical means for ultra-high density and ultra-fast magnetic storage.