王文敏-凯发k8国际首页登录

prof. wang wenmin

school of computer science and engineering, faculty of innovation engineering

macau university of science and technology

office: a320

e-mail:

• ph.d. in computer science and technology, harbin institute of technology, china, 1989.
• m.s. in computer science and technology, harbin institute of technology, china, 1986.
• b.s. in computer science and technology, harbin university of science and technology, china, 1983.

• artificial intelligence
• machine learning

• computer vision
• multimedia retrieval
• artificial intelligent and machine learning

• mar 2022 - present professor/program director, school of computer science and engineering, macau university of science and technology, macau.
• jan 2019 - feb 2022 professor/program director, international institute for next generation internet, macau university of science and technology, macau.
• jan 2010 - dec 2018 professor/executive vice dean/dean, school of electronic and computer engineering, peking university, china. deputy director, guangzhou r&d center, national engineering lab for video technology, china.
• mar 2003 - dec 2009 senior engineer/general manager, dtv department, zentek technology japan, inc., japan.
• aug 2000 - feb 2003 senior engineer/chief engineer, liberate technologies inc. usa.
• jun 1992 - jul 2000 section vice chief/staff engineer, canotec co., inc., canon group. japan.
• mar 1989 - may 1991 assistant professor/associate professor, harbin university of science and technology/harbin institute of technology, china.

[1] cheng yu, wenmin wang, honglei li, and roberto bugiolacchi. ""fast 2-step regularization on style optimization for real face morphing ."" neural networks, accepted, in aug. 2022.

[2] shixiong zhang, wenmin wang, honglei li, and shenyong zhang. ""evtracker: an event-driven spatiotemporal method for dynamic object tracking."" sensors, 22(16), 2022.

[3] jianhao yan, wenmin wang, cheng yu. “affective word embedding in affective explanation generation for fine art paintings.” pattern recognition letters, 161, 2022.

[4] honglei li, wenmin wang, cheng yu, and shixiong zhang. “swapinpaint: identity-specific face inpainting with identity swapping.” ieee trans. on circuits and systemsfor video technology (tcsvt)," "32(7), 2022.

[5] cheng yu, and wenmin wang. “fast transformation of discriminators into encoders using pre-trained gans.” pattern recognition letters, 153, 2022.

[6] cheng yu, wenmin wang, and jianhao yan. ""self-supervised animation synthesis through adversarial training."" ieee access, 8, 2020.

[7] xiongtao chen, and wenmin wang. “uni-and-bi-directional video prediction via learning object-centric transformation.” ieee trans. on multimedia (tmm), 22(6), 2020.

[8] jinzhuo wang, wenmin wang, and wen gao. “fast and accurate action detection in videos with motion-centric attention model. ” ieee trans. on circuits and systemsfor video technology (tcsvt),

30(1), 2020.

[9] lun huang, wenmin wang, yaxian xia, and jie chen. “adaptively aligned image captioning via adaptive attention time.” advances in neural information processing systems (neurips), dec 2019.

vancouver, canada.

[10] jinzhuo wang, wenmin wang, and wen gao. “predicting diverse future frames with local transformation-guided masking.” ieee trans. on circuits and systemsfor video technology (tcsvt), 29(12), 2019.

[11] lun huang, wenmin wang, jie chen, and xiaoyong wei. “attention on attention for image captioning.” ieee int’l conf. on computer vision (iccv), oct 2019. seoul, korea.

[12] 王文敏，《人工智能原理》，高等教育出版社， 2019年8月(550千字)。

[13] qian wu, wenmin wang, xiongtao chen, weimian li. “video prediction with temporal-spatial attention mechanism and deep perceptual similarity branch.” ieee int’l conf. on multimedia and expo (icme), jul 2019. shanghai, china.

[14] zheng yu, wenmin wang, ge li. “multi-step self-attention network for cross-modal retrieval based on a limited text space.” ieee int’l conf. on acoustics, speech, and signal processing (icassp), may 2019. brighton, uk.

[15] jinzhuo wang, wenmin wang, and wen gao. “multiscale deep alternative neural network for large scale video classification.” ieee trans. on multimedia (tmm), 20(10), 2018.

[16] ruonan zhang, and wenmin wang. “second- and high-order graph matching for correspondence problem.” ieee trans. on circuits and systemsfor video technology (tcsvt), 28(10), 2018.

[17] mengdi fan, wenmin wang, peilei dong. liang han, ronggang wang, and ge li. “cross-media retrieval by learning rich semantic embeddings of multimedia.” acmint’l conf. on multimedia (acm mm), oct 2017. mountain view, california, usa.

[18] baoyang chen, wenmin wang, and jinzhuo wang. “video imagination from a single image with transformation generation.” thematic workshops, acm int’l conf. on multimedia (acm mm), oct 2017. mountain view, california, usa.

[19] weimian li, wenmin wang, xiongtao chen, jinzhuo wang, and ge li. ""a joint model for action localization and classification in untrimmed video with visual attention"", ieee int’l conf. on multimedia and expo (icme), jul 2017. hong kong, china.

[20] hongmeng song, wenmin wang, jinzhuo wang, and ronggang wang. “collaborative deep networks for pedestrian detection.” ieee int’l conf. on multimedia big data (bigmm), apr 2017. laguna hills, california, usa. (best paper award).

[21] jinzhuo wang, wenmin wang, ronggang wang, and wen gao. “beyond monte carlo tree search: playing go with deep alternative neural network and long-term evaluation.” association for the advance ofartificial intelligence (aaai), feb 2017. san francisco, usa.

[22] jinzhuo wang, wenmin wang, xiongtao chen, ronggang wang, and wen gao. “deep alternative neural network: exploring contexts as early as possible for action recognition .” advances in neural information processing systems (nips), dec 2016. barcelona, spain.

[23] yang zhao, ronggang wang, wenmin wang, and wen gao. “multilevel modified finite radon transform network for image upsampling,” ieee trans. on circuits and systems for video technology (tcsvt), 26(12), 2016."

[24] peilei dong, and wenmin wang. “better region proposals for pedestrian detection with r-cnn.” ieee int’l conf. on visual communications and image processing (vcip), nov 2016. chengdu, china. (best top 10% paper award)

[25] jinzhuo wang, wenmin wang, ronggang wang, and wen gao. “csps: an adaptive pooling method for image classification. ” ieee trans. on multimedia (tmm), 18(6) , 2016.

[26] yang zhang, yunqing xia, yi liu, and wenmin wang. “clustering sentences with density peaks for multi-document summarization. ” the 2015 annual conf. of the north american chapter of the acl (naacl), oct 2015. denver, usa.

[27] mingmin zhen, wenmin wang, and ronggang wang. regional subspace projection coding for image retrieval, acm int’l conf. on multimedia retrieval (icmr), jun 2016. new york, usa.

[28] yang zhao, ronggang wang, wenmin wang, and wen gao. “high resolution local structure constrained image upsampling.” ieee trans. on image processing (tip), 24(11) , 2015."

[1] wenmin wang, lun ran, ronggang wang, shengfu dong, zhenyu wang, ying li, hui zhao, and wen gao. “image text bidirectional retrieval method based on multi view joint embedding space.” china patent 201710545632.x.

[2] wenmin wang, hongmeng song, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “collaborative deep networks for pedestrian detection.” china patent 201611166903.2; us patent 10,867,167.

[3] wenmin wang, liang han, ronggang wang, shengfu dong, zhenyu wang, ying li , and wen gao. “cross-media retrieval method.” china patent 201610544156.5; us patent 10,719,664.

[4] wenmin wang, mengdi fan, ronggang wang, shengfu dong, zhenyu wang, ying li , and wen gao. “cross-media retrieval method based on deep semantic space.” china patent 201610398342.2.

[5] wenmin wang, ruonan zhang, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “method and a device for image matching.” china patent 201610022261.2; us patent 10,699,156.

[6] wenmin wang, mingmin zhen, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “image representation method and processing device based on local pca whiting.” china patent 201510585376.8; us patent 10,424,052.

[7] wenmin wang, mingmin zhen, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “method and a device for extracting local features of a three-dimensional point cloud.” china patent 201510340617.2; us patent 10,339,409.

[8] wenmin wang, qinshui chen, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “method and its device for stereo 3d web.” china patent 201510202663.6.

[9] wenmin wang, jianlong zhang, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “method and its system for rendering stereo 3d graph.” china patent 201410302221.4.

[10] wenmin wang, ronggang wang, shengfu dong, zhenyu wang, ying li, and wen gao. “method and system for realizing stereo 3d web pages.” china patent 201310290469.9.

[11] wenmin wang, xiaozhen wang, wen gao, and fan liang. “a implement method for stereo 3d web browser based on principle of binocular stereoscopic vision.” china patent 201110131554.1.

senior member of ieee/ccf, member of acm/caai.