https://nkdailab.github.io/https://nkdailab.github.io/research/https://nkdailab.github.io/team/https://nkdailab.github.io/publications/https://nkdailab.github.io/achievements/https://nkdailab.github.io/activities/https://nkdailab.github.io/news/https://nkdailab.github.io/join/https://nkdailab.github.io/search/https://nkdailab.github.io/gallery/https://nkdailab.github.io/research/areas/https://nkdailab.github.io/research/group/https://nkdailab.github.io/research/results/https://nkdailab.github.io/team/zhihai-he/https://nkdailab.github.io/team/qifan-liu/https://nkdailab.github.io/team/weiming-chen/https://nkdailab.github.io/team/junyang-ji/https://nkdailab.github.io/team/zhineng-zhao/https://nkdailab.github.io/team/fan-du/https://nkdailab.github.io/team/yongyong-gao/https://nkdailab.github.io/team/zhiwei-hu/https://nkdailab.github.io/team/yanji-huang/https://nkdailab.github.io/team/bowen-liao/https://nkdailab.github.io/team/siyi-liu/https://nkdailab.github.io/team/ziqiong-liu/https://nkdailab.github.io/team/kaize-luo/https://nkdailab.github.io/team/yangxing-shang/https://nkdailab.github.io/team/bingyan-wu/https://nkdailab.github.io/team/ke-yang/https://nkdailab.github.io/team/zihao-zhou/https://nkdailab.github.io/team/xiyuan-liu/https://nkdailab.github.io/team/weiye-qian/https://nkdailab.github.io/team/yijia-wang/https://nkdailab.github.io/team/jingjun-xu/https://nkdailab.github.io/team/zhihan-zhu/https://nkdailab.github.io/team/xinhua-fan/https://nkdailab.github.io/team/youming-li/https://nkdailab.github.io/team/siqi-wu/https://nkdailab.github.io/team/qingnan-chen/https://nkdailab.github.io/team/shuoshuo-chen/https://nkdailab.github.io/team/bowen-hai/https://nkdailab.github.io/team/junyi-he/https://nkdailab.github.io/team/xueting-hu/https://nkdailab.github.io/team/jiyuan-jia/https://nkdailab.github.io/team/zhehan-kan/https://nkdailab.github.io/team/jing-li/https://nkdailab.github.io/team/ruoyi-li/https://nkdailab.github.io/team/ying-shen/https://nkdailab.github.io/team/yushun-tang/https://nkdailab.github.io/team/heng-xu/https://nkdailab.github.io/team/ke-yu/https://nkdailab.github.io/team/ce-zhang/https://nkdailab.github.io/team/yi-zhang/https://nkdailab.github.io/team/yangyang-sun/https://nkdailab.github.io/team/jiaqi-lian/https://nkdailab.github.io/team/zixuan-lv/https://nkdailab.github.io/team/chenghu-qiu/https://nkdailab.github.io/team/zihao-huang/https://nkdailab.github.io/team/zishan-zhou/https://nkdailab.github.io/team/guanlin-chen/https://nkdailab.github.io/publications/controllable-generative-reference-for-stereo-image-compression-via-reliability-aware-gating/https://nkdailab.github.io/publications/omniportrait-fine-grained-personalized-portrait-synthesis-via-pivotal-optimization/https://nkdailab.github.io/publications/omnicvr-a-benchmark-for-omni-composed-video-retrieval-with-vision-audio-and-text/https://nkdailab.github.io/publications/training-free-dual-hyperbolic-adapters-for-better-cross-modal-reasoning/https://nkdailab.github.io/publications/rasa-reliability-aware-selective-adaptation-via-hybrid-reliability-threshold/https://nkdailab.github.io/publications/deeply-conditioned-image-compression-via-self-generated-priors/https://nkdailab.github.io/publications/causallens-sensitivity-guided-multi-head-causal-intervention-for-hallucination-mitigation-in-large-vision-language-models/https://nkdailab.github.io/publications/audio-driven-multi-modal-unobtrusive-health-monitoring-and-inference-for-smart-eldercare-at-home/https://nkdailab.github.io/publications/latentedit-adaptive-latent-control-for-consistent-semantic-editing/https://nkdailab.github.io/publications/enhancing-visual-and-semantic-alignment-of-multimodal-large-models-in-medical-images/https://nkdailab.github.io/publications/open-vocabulary-multi-label-classification-based-on-cross-modal-feature-reconstruction-learning/https://nkdailab.github.io/publications/cross-modal-few-shot-learning-with-second-order-neural-ordinary-differential-equations/https://nkdailab.github.io/publications/conditional-latent-coding-with-learnable-synthesized-reference-for-deep-image-compression/https://nkdailab.github.io/publications/window-based-channel-attention-for-wavelet-enhanced-learned-image-compression/https://nkdailab.github.io/publications/dual-path-adversarial-lifting-for-domain-shift-correction-in-online-test-time-adaptation/https://nkdailab.github.io/publications/conceptual-codebook-learning-for-vision-language-models/https://nkdailab.github.io/publications/training-free-feature-reconstruction-with-sparse-optimization-for-vision-language-models/https://nkdailab.github.io/publications/domain-conditioned-transformer-for-fully-test-time-adaptation/https://nkdailab.github.io/publications/learning-inference-time-drift-sensor-actuator-for-domain-generalization/https://nkdailab.github.io/publications/cross-constrained-progressive-inference-for-3d-hand-pose-estimation-with-dynamic-observer-decision-adjuster-networks/https://nkdailab.github.io/publications/concept-guided-prompt-learning-for-generalization-in-vision-language-models/https://nkdailab.github.io/publications/learning-to-adapt-clip-for-few-shot-monocular-depth-estimation/https://nkdailab.github.io/publications/unsupervised-action-recognition-using-spatiotemporal-adaptive-and-attention-guided-refining-network/https://nkdailab.github.io/publications/learning-visual-conditioning-tokens-to-correct-domain-shift-for-fully-test-time-adaptation/https://nkdailab.github.io/publications/hierarchical-spatial-temporal-masked-contrast-for-skeleton-action-recognition/https://nkdailab.github.io/publications/full-coverage-unobtrusive-health-monitoring-of-elders-at-homes/https://nkdailab.github.io/publications/cross-modal-concept-learning-and-inference-for-vision-language-models/https://nkdailab.github.io/publications/critical-sampling-for-robust-evolution-operator-learning-of-unknown-dynamical-systems/https://nkdailab.github.io/publications/unsupervised-prototype-adapter-for-vision-language-models/https://nkdailab.github.io/publications/cross-inferential-networks-for-source-free-unsupervised-domain-adaptation/https://nkdailab.github.io/publications/self-correctable-and-adaptable-inference-for-generalizable-human-pose-estimation/https://nkdailab.github.io/publications/neuro-modulated-hebbian-learning-for-fully-test-time-adaptation/https://nkdailab.github.io/publications/cycle-optimization-metric-learning-for-few-shot-classification/https://nkdailab.github.io/publications/critical-sampling-for-data-driven-modeling-of-unknown-dynamical-systems/https://nkdailab.github.io/publications/bdc-adapter-brownian-distance-covariance-for-better-vision-language-reasoning/https://nkdailab.github.io/publications/self-constrained-inference-optimization-on-structural-groups-for-human-pose-estimation/https://nkdailab.github.io/publications/geometric-machine-learning-research-and-applications/https://nkdailab.github.io/publications/self-guided-information-for-few-shot-classification/https://nkdailab.github.io/publications/local-semantic-correlation-modeling-over-graph-neural-networks-for-deep-feature-embedding-and-image-retrieval/https://nkdailab.github.io/publications/jointly-learnt-networks-for-future-action-anticipation-via-self-knowledge-distillation-and-cycle-consistency/https://nkdailab.github.io/publications/contrastive-bayesian-analysis-for-deep-metric-learning/https://nkdailab.github.io/publications/coded-residual-transform-for-generalizable-deep-metric-learning/https://nkdailab.github.io/news/2025.3.6/https://nkdailab.github.io/news/2025.2.25/https://nkdailab.github.io/news/zhehan-kan-a-postgraduate-of-zhihai-he-research-group-of-the-department-of-electronics-in-occlusion-and-generalization-issue-in-human-pose-estimation-with-a-novel-self-constrained-approach/https://nkdailab.github.io/en/https://nkdailab.github.io/en/research/https://nkdailab.github.io/en/team/https://nkdailab.github.io/en/publications/https://nkdailab.github.io/en/achievements/https://nkdailab.github.io/en/activities/https://nkdailab.github.io/en/news/https://nkdailab.github.io/en/join/https://nkdailab.github.io/en/search/https://nkdailab.github.io/en/gallery/https://nkdailab.github.io/en/research/areas/https://nkdailab.github.io/en/research/group/https://nkdailab.github.io/en/research/results/https://nkdailab.github.io/en/team/zhihai-he/https://nkdailab.github.io/en/team/qifan-liu/https://nkdailab.github.io/en/team/weiming-chen/https://nkdailab.github.io/en/team/junyang-ji/https://nkdailab.github.io/en/team/zhineng-zhao/https://nkdailab.github.io/en/team/fan-du/https://nkdailab.github.io/en/team/yongyong-gao/https://nkdailab.github.io/en/team/zhiwei-hu/https://nkdailab.github.io/en/team/yanji-huang/https://nkdailab.github.io/en/team/bowen-liao/https://nkdailab.github.io/en/team/siyi-liu/https://nkdailab.github.io/en/team/ziqiong-liu/https://nkdailab.github.io/en/team/kaize-luo/https://nkdailab.github.io/en/team/yangxing-shang/https://nkdailab.github.io/en/team/bingyan-wu/https://nkdailab.github.io/en/team/ke-yang/https://nkdailab.github.io/en/team/zihao-zhou/https://nkdailab.github.io/en/team/xiyuan-liu/https://nkdailab.github.io/en/team/weiye-qian/https://nkdailab.github.io/en/team/yijia-wang/https://nkdailab.github.io/en/team/jingjun-xu/https://nkdailab.github.io/en/team/zhihan-zhu/https://nkdailab.github.io/en/team/xinhua-fan/https://nkdailab.github.io/en/team/youming-li/https://nkdailab.github.io/en/team/siqi-wu/https://nkdailab.github.io/en/team/qingnan-chen/https://nkdailab.github.io/en/team/shuoshuo-chen/https://nkdailab.github.io/en/team/bowen-hai/https://nkdailab.github.io/en/team/junyi-he/https://nkdailab.github.io/en/team/xueting-hu/https://nkdailab.github.io/en/team/jiyuan-jia/https://nkdailab.github.io/en/team/zhehan-kan/https://nkdailab.github.io/en/team/jing-li/https://nkdailab.github.io/en/team/ruoyi-li/https://nkdailab.github.io/en/team/ying-shen/https://nkdailab.github.io/en/team/yushun-tang/https://nkdailab.github.io/en/team/heng-xu/https://nkdailab.github.io/en/team/ke-yu/https://nkdailab.github.io/en/team/ce-zhang/https://nkdailab.github.io/en/team/yi-zhang/https://nkdailab.github.io/en/team/yangyang-sun/https://nkdailab.github.io/en/team/jiaqi-lian/https://nkdailab.github.io/en/team/zixuan-lv/https://nkdailab.github.io/en/team/chenghu-qiu/https://nkdailab.github.io/en/team/zihao-huang/https://nkdailab.github.io/en/team/zishan-zhou/https://nkdailab.github.io/en/team/guanlin-chen/https://nkdailab.github.io/en/publications/controllable-generative-reference-for-stereo-image-compression-via-reliability-aware-gating/https://nkdailab.github.io/en/publications/omniportrait-fine-grained-personalized-portrait-synthesis-via-pivotal-optimization/https://nkdailab.github.io/en/publications/omnicvr-a-benchmark-for-omni-composed-video-retrieval-with-vision-audio-and-text/https://nkdailab.github.io/en/publications/training-free-dual-hyperbolic-adapters-for-better-cross-modal-reasoning/https://nkdailab.github.io/en/publications/rasa-reliability-aware-selective-adaptation-via-hybrid-reliability-threshold/https://nkdailab.github.io/en/publications/deeply-conditioned-image-compression-via-self-generated-priors/https://nkdailab.github.io/en/publications/causallens-sensitivity-guided-multi-head-causal-intervention-for-hallucination-mitigation-in-large-vision-language-models/https://nkdailab.github.io/en/publications/audio-driven-multi-modal-unobtrusive-health-monitoring-and-inference-for-smart-eldercare-at-home/https://nkdailab.github.io/en/publications/latentedit-adaptive-latent-control-for-consistent-semantic-editing/https://nkdailab.github.io/en/publications/enhancing-visual-and-semantic-alignment-of-multimodal-large-models-in-medical-images/https://nkdailab.github.io/en/publications/open-vocabulary-multi-label-classification-based-on-cross-modal-feature-reconstruction-learning/https://nkdailab.github.io/en/publications/cross-modal-few-shot-learning-with-second-order-neural-ordinary-differential-equations/https://nkdailab.github.io/en/publications/conditional-latent-coding-with-learnable-synthesized-reference-for-deep-image-compression/https://nkdailab.github.io/en/publications/window-based-channel-attention-for-wavelet-enhanced-learned-image-compression/https://nkdailab.github.io/en/publications/dual-path-adversarial-lifting-for-domain-shift-correction-in-online-test-time-adaptation/https://nkdailab.github.io/en/publications/conceptual-codebook-learning-for-vision-language-models/https://nkdailab.github.io/en/publications/training-free-feature-reconstruction-with-sparse-optimization-for-vision-language-models/https://nkdailab.github.io/en/publications/domain-conditioned-transformer-for-fully-test-time-adaptation/https://nkdailab.github.io/en/publications/learning-inference-time-drift-sensor-actuator-for-domain-generalization/https://nkdailab.github.io/en/publications/cross-constrained-progressive-inference-for-3d-hand-pose-estimation-with-dynamic-observer-decision-adjuster-networks/https://nkdailab.github.io/en/publications/concept-guided-prompt-learning-for-generalization-in-vision-language-models/https://nkdailab.github.io/en/publications/learning-to-adapt-clip-for-few-shot-monocular-depth-estimation/https://nkdailab.github.io/en/publications/unsupervised-action-recognition-using-spatiotemporal-adaptive-and-attention-guided-refining-network/https://nkdailab.github.io/en/publications/learning-visual-conditioning-tokens-to-correct-domain-shift-for-fully-test-time-adaptation/https://nkdailab.github.io/en/publications/hierarchical-spatial-temporal-masked-contrast-for-skeleton-action-recognition/https://nkdailab.github.io/en/publications/full-coverage-unobtrusive-health-monitoring-of-elders-at-homes/https://nkdailab.github.io/en/publications/cross-modal-concept-learning-and-inference-for-vision-language-models/https://nkdailab.github.io/en/publications/critical-sampling-for-robust-evolution-operator-learning-of-unknown-dynamical-systems/https://nkdailab.github.io/en/publications/unsupervised-prototype-adapter-for-vision-language-models/https://nkdailab.github.io/en/publications/cross-inferential-networks-for-source-free-unsupervised-domain-adaptation/https://nkdailab.github.io/en/publications/self-correctable-and-adaptable-inference-for-generalizable-human-pose-estimation/https://nkdailab.github.io/en/publications/neuro-modulated-hebbian-learning-for-fully-test-time-adaptation/https://nkdailab.github.io/en/publications/cycle-optimization-metric-learning-for-few-shot-classification/https://nkdailab.github.io/en/publications/critical-sampling-for-data-driven-modeling-of-unknown-dynamical-systems/https://nkdailab.github.io/en/publications/bdc-adapter-brownian-distance-covariance-for-better-vision-language-reasoning/https://nkdailab.github.io/en/publications/self-constrained-inference-optimization-on-structural-groups-for-human-pose-estimation/https://nkdailab.github.io/en/publications/geometric-machine-learning-research-and-applications/https://nkdailab.github.io/en/publications/self-guided-information-for-few-shot-classification/https://nkdailab.github.io/en/publications/local-semantic-correlation-modeling-over-graph-neural-networks-for-deep-feature-embedding-and-image-retrieval/https://nkdailab.github.io/en/publications/jointly-learnt-networks-for-future-action-anticipation-via-self-knowledge-distillation-and-cycle-consistency/https://nkdailab.github.io/en/publications/contrastive-bayesian-analysis-for-deep-metric-learning/https://nkdailab.github.io/en/publications/coded-residual-transform-for-generalizable-deep-metric-learning/https://nkdailab.github.io/en/news/2025.3.6/https://nkdailab.github.io/en/news/2025.2.25/https://nkdailab.github.io/en/news/zhehan-kan-a-postgraduate-of-zhihai-he-research-group-of-the-department-of-electronics-in-occlusion-and-generalization-issue-in-human-pose-estimation-with-a-novel-self-constrained-approach/