Xiaogang Cheng

Guolei Sun, Xiaogang Cheng, Zhaochong An et al. · microsoft-research

Recently, indiscernible/camouflaged scene understanding has attracted lots of research attention in the vision community. We further advance the frontier of this field by systematically studying a new challenge named indiscernible object counting (IOC), the goal of which is to count objects that are blended with respect to their surroundings. Due to a lack of appropriate IOC datasets, we present a large-scale dataset IOCfish5K which contains a total of 5,637 high-resolution images and 659,024 annotated center points. Our dataset consists of a large number of indiscernible objects (mainly fish) in underwater scenes, making the annotation process all the more challenging. IOCfish5K is superior to existing datasets with indiscernible scenes because of its larger scale, higher image resolutions, more annotations, and denser scenes. All these aspects make it the most challenging dataset for IOC so far, supporting progress in this area. Benefiting from the recent advancements of depth estimation foundation models, we construct high-quality depth maps for IOCfish5K by generating pseudo labels using the Depth Anything V2 model. The RGB-D version of IOCfish5K is named IOCfish5K-D. For benchmarking purposes on IOCfish5K, we select 14 mainstream methods for object counting and carefully evaluate them. For multimodal IOCfish5K-D, we evaluate other 4 popular multimodal counting methods. Furthermore, we propose IOCFormer, a new strong baseline that combines density and regression branches in a unified framework and can effectively tackle object counting under concealed scenes. We also propose IOCFormer-D to enable the effective usage of depth modality in helping detect and count objects hidden in their environments. Experiments show that IOCFormer and IOCFormer-D achieve state-of-the-art scores on IOCfish5K and IOCfish5K-D, respectively.

Jie Qin, Peng Zheng, Yichao Yan et al.

Person search aims to jointly localize and identify a query person from natural, uncropped images, which has been actively studied over the past few years. In this paper, we delve into the rich context information globally and locally surrounding the target person, which we refer to as scene and group context, respectively. Unlike previous works that treat the two types of context individually, we exploit them in a unified global-local context network (GLCNet) with the intuitive aim of feature enhancement. Specifically, re-ID embeddings and context features are simultaneously learned in a multi-stage fashion, ultimately leading to enhanced, discriminative features for person search. We conduct the experiments on two person search benchmarks (i.e., CUHK-SYSU and PRW) as well as extend our approach to a more challenging setting (i.e., character search on MovieNet). Extensive experimental results demonstrate the consistent improvement of the proposed GLCNet over the state-of-the-art methods on all three datasets. Our source codes, pre-trained models, and the new dataset are publicly available at: https://github.com/ZhengPeng7/GLCNet.

Xiaogang Cheng, Bin Yang, Kaige Tan et al.

In human-centered intelligent building, real-time measurements of human thermal comfort play critical roles and supply feedback control signals for building heating, ventilation, and air conditioning (HVAC) systems. Due to the challenges of intra- and inter-individual differences and skin subtleness variations, there is no satisfactory solution for thermal comfort measurements until now. In this paper, a non-invasive measuring method based on skin sensitivity index and deep learning (NISDL) was proposed to measure real-time skin temperature. A new evaluating index, named skin sensitivity index (SSI), was defined to overcome individual differences and skin subtleness variations. To illustrate the effectiveness of SSI proposed, two multi-layers deep learning framework (NISDL method I and II) was designed and the DenseNet201 was used for extracting features from skin images. The partly personal saturation temperature (NIPST) algorithm was use for algorithm comparisons. Another deep learning algorithm without SSI (DL) was also generated for algorithm comparisons. Finally, a total of 1.44 million image data was used for algorithm validation. The results show that 55.6180% and 52.2472% error values (NISDL method I, II) are scattered at [0, 0.25), and the same error intervals distribution of NIPST is 35.3933%.

Bin Yang, Xiaogang Cheng, Dengxin Dai et al.

Individual thermal comfort perception gives important feedback signals for energy efficient control of smart buildings. However, there is no effective method to measure real-time thermal comfort status of individual occupant until now. For overcoming this challenge, a novel macro posed based non-invasive perception method for thermal comfort (NIMAP) was presented. The occupant pose images were captured by normal phone camera (computer or cell phone) and the corresponding 2D coordinates can be obtained. Based on this, a novel pose recognition algorithm for thermal comfort, including 12 sub-algorithms, was presented. The 12 thermal comfort related macro poses can be recognized. Further, based on Fanger theory, 369 subjects were invited for subjective questionnaire survey. 3 human occupants participated in the validation of the proposed method and massive data were collected. All the 12 thermal comfort related poses can be recognized effectively.

Xiaogang Cheng, Bin Yang, Anders Hedman et al.

Human thermal comfort measurement plays a critical role in giving feedback signals for building energy efficiency. A non-invasive measuring method based on subtleness magnification and deep learning (NIDL) was designed to achieve a comfortable, energy efficient built environment. The method relies on skin feature data, e.g., subtle motion and texture variation, and a 315-layer deep neural network for constructing the relationship between skin features and skin temperature. A physiological experiment was conducted for collecting feature data (1.44 million) and algorithm validation. The non-invasive measurement algorithm based on a partly-personalized saturation temperature model (NIPST) was used for algorithm performance comparisons. The results show that the mean error and median error of the NIDL are 0.4834 Celsius and 0.3464 Celsius which is equivalent to accuracy improvements of 16.28% and 4.28%, respectively.

Xiaogang Cheng, Guoqing Liu, Anders Hedman et al.

Vision-based methods for visibility estimation can play a critical role in reducing traffic accidents caused by fog and haze. To overcome the disadvantages of current visibility estimation methods, we present a novel data-driven approach based on Gaussian image entropy and piecewise stationary time series analysis (SPEV). This is the first time that Gaussian image entropy is used for estimating atmospheric visibility. To lessen the impact of landscape and sunshine illuminance on visibility estimation, we used region of interest (ROI) analysis and took into account relative ratios of image entropy, to improve estimation accuracy. We assume fog and haze cause blurred images and that fog and haze can be considered as a piecewise stationary signal. We used piecewise stationary time series analysis to construct the piecewise causal relationship between image entropy and visibility. To obtain a real-world visibility measure during fog and haze, a subjective assessment was established through a study with 36 subjects who performed visibility observations. Finally, a total of two million videos were used for training the SPEV model and validate its effectiveness. The videos were collected from the constantly foggy and hazy Tongqi expressway in Jiangsu, China. The contrast model of visibility estimation was used for algorithm performance comparison, and the validation results of the SPEV model were encouraging as 99.14% of the relative errors were less than 10%.