Liang Li

Liang Li

Auto-generated profile, based on publicly available information

Liang Li

Beijing Institute of Basic Medical Sciences, Beijing, China

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 December 2021 in PLOS Biology
Verified
Journal Article
Open access
PDF file
CHChuanliang Han
Chuanliang Han
TWTian Wang
Tian Wang
YYYi Yang
Yi Yang
YWYujie Wu
Yujie Wu
YLYang Li
Yang Li
DXDajun Xing
Dajun Xing
Gamma rhythms in many brain regions, including the primary visual cortex (V1), are thought to play a role in information processing. Here, we report a surprising finding of 3 narrowband gamma rhythms in V1 that processed distinct spatial frequency (SF) signals and had different neural origins. The low gamma (LG; 25 to 40 Hz) rhythm was generated at the V1 superficial layer and preferred a higher SF compared with spike activity, whereas both the medium gamma (MG; 40 to 65 Hz), generated at the cortical level, and the high gamma HG; (65 to 85 Hz), originated precortically, preferred lower SF information. Furthermore, compared with the rates of spike activity, the powers of the 3 gammas had better performance in discriminating the edge and surface of simple objects. These findings suggest that gamma rhythms reflect the neural dynamics of neural circuitries that process different SF information in the visual system, which may be crucial for multiplexing SF information and synchronizing different features of an object.

8 September 2021 in Science Translational Medicine
Verified
Journal Article
Closed access
YWYuqing Wang
Yuqing Wang
SLShanshan Lu
Shanshan Lu
YCYifei Chen
Yifei Chen
LLLiang Li
Liang Li
XLXia Li
Xia Li
YWYi-Zheng Wang
Yi-Zheng Wang
Activation of SHH pathway contributes to glutamate toxicity, and inhibition of SMO reduces ischemic brain damage in mice and monkeys.

11 July 2022 in Journal of Vision
Verified
Journal Article
Open access
PDF file
QZQi Zhang
Qi Zhang
ZHZhibang Huang
Zhibang Huang
LLLiang Li
Liang Li
SLSheng Li
Sheng Li
Training serves as an effective approach to improve visual search performance when the target does not automatically pop out from the distractors. In the present study, we trained participants on a conjunction visual search task and examined the training effects in behavior and eye movement. The results of Experiments 1 to 4 showed that training improved behavioral performance and reduced the number of saccades and overall scanning time. Training also increased the search initiation time before the first saccade and the proportion of trials in which the participants correctly identified the target without any saccade, but these effects were modulated by stimulus’ parameters. In Experiment 5, we simultaneously recorded eye movements and electroencephalography signals and the results revealed significant N2 posterior contralateral (N2pc) components after the stimulus onset (i.e., stimulus-locked) and before the first saccade (i.e., saccade-locked) when the search target was the trained one. These N2pc components can be considered as the neural signatures for the enhanced covert attention to the trained target. Together with the training-induced increase in functional visual field, these mechanisms could support the beneficial effects of increased search initiation time and reduced number of saccades. These findings suggest that visual search training enhanced covert attention to target and optimized overt eye movements to facilitate search performance.

13 February 2025 in International Journal of Computer Vision
Verified
Journal Article
Open access
PDF file
JGJin Gao
Jin Gao
SLShubo Lin
Shubo Lin
SWShaoru Wang
Shaoru Wang
YKYutong Kou
Yutong Kou
ZLZeming Li
Zeming Li
WHWeiming Hu
Weiming Hu
Masked image modeling (MIM) pre-training for large-scale vision transformers (ViTs) has enabled promising downstream performance on top of the learned self-supervised ViT features. In this paper, we question if the extremely simple lightweight ViTs’ fine-tuning performance can also benefit from this pre-training paradigm, which is considerably less studied yet in contrast to the well-established lightweight architecture design methodology. We use an observation-analysis-solution flow for our study. We first systematically observe different behaviors among the evaluated pre-training methods with respect to the downstream fine-tuning data scales. Furthermore, we analyze the layer representation similarities and attention maps across the obtained models, which clearly show the inferior learning of MIM pre-training on higher layers, leading to unsatisfactory transfer performance on data-insufficient downstream tasks. This finding is naturally a guide to designing our distillation strategies during pre-training to solve the above deterioration problem. Extensive experiments have demonstrated the effectiveness of our approach. Our pre-training with distillation on pure lightweight ViTs with vanilla/hierarchical design (5.7M/6.5M) can achieve \(79.4\%\)/\(78.9\%\) top-1 accuracy on ImageNet-1K. It also enables SOTA performance on the ADE20K segmentation task (\(42.8\%\) mIoU) and LaSOT tracking task (\(66.1\%\) AUC) in the lightweight regime. The latter even surpasses all the current SOTA lightweight CPU-realtime trackers.

7 March 2022 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
JGJin Gao
Jin Gao
YLYan Lu
Yan Lu
XQXiaojuan Qi
Xiaojuan Qi
YKYutong Kou
Yutong Kou
BLBing Li
Bing Li
WHWeiming Hu
Weiming Hu
Tracking visual objects from a single initial exemplar in the testing phase has been broadly cast as a one-/few-shot problem, i.e., one-shot learning for initial adaptation and few-shot learning for online adaptation. The recent few-shot online adaptation methods incorporate the prior knowledge from large amounts of annotated training data via complex meta-learning optimization in the offline phase. This helps the online deep trackers to achieve fast adaptation and reduce overfitting risk in tracking. In this paper, we propose a simple yet effective recursive least-squares estimator-aided online learning approach for few-shot online adaptation without requiring offline training. It allows an in-built memory retention mechanism for the model to remember the knowledge about the object seen before, and thus the seen data can be safely removed from training. This also bears certain similarities to the emerging continual learning field in preventing catastrophic forgetting. This mechanism enables us to unveil the power of modern online deep trackers without incurring too much extra computational cost. We evaluate our approach based on two networks in the online learning families for tracking, i.e., multi-layer perceptrons in RT-MDNet and convolutional neural networks in DiMP. The consistent improvements on several challenging tracking benchmarks demonstrate its effectiveness and efficiency.

17 August 2023 in iScience
Verified
Journal Article
Open access
YHYifei Han
Yifei Han
WHWenhao Han
Wenhao Han
LLLiang Li
Liang Li
TZTao Zhang
Tao Zhang
YWYizheng Wang
Yizheng Wang
Humans can distinguish flying birds from drones based solely on motion features when no image information is available. However, it remains unclear which motion features of animate motion induce our animacy perception. To address this, we first analyzed the differences in centroid motion between birds and drones, and discovered that birds exhibit greater acceleration, angular speed, and trajectory fluctuations. We further determined the order of their importance in evoking animacy perception was trajectory fluctuations, acceleration, and speed. More interestingly, people judge whether a moving object is alive using a feature-matching strategy, implying that animacy perception is induced in a key feature-triggered way rather than relying on the accumulation of evidence. Our findings not only shed light on the critical motion features that induce animacy perception and their relative contributions but also have important implications for developing target classification algorithms based on motion features.