We present GEM, a Generalizable Ego-vision Multimodal world model that predicts future frames using a reference frame, sparse features, human poses, and ego-trajectories. Hence, our model has precise control over object dynamics, ego-agent motion and human poses. GEM generates paired RGB and depth outputs for richer spatial understanding. We introduce autoregressive noise schedules to enable stable long-horizon generations. Our dataset is comprised of 4000+ hours of multimodal data across domains like autonomous driving, egocentric human activities, and drone flights. Pseudo-labels are used to get depth maps, ego-trajectories, and human poses. We use a comprehensive evaluation framework, including a new Control of Object Manipulation (COM) metric, to assess controllability. Experiments show GEM excels at generating diverse, controllable scenarios and temporal consistency over long generations. Code, models, and datasets are fully open-sourced1.

A deep-learning-based cloud detection and classification algorithm for advanced Himawari imager (AHI) measurements from the geostationary satellite Himawari-8 has been developed. It is found that a combination of observed radiances and simulated clear-sky radiances can substantially improve cloud phase discrimination, especially for optically thin clouds. Therefore, cloud detection, cloud phase classification, and multilayer cloud detection are obtained simultaneously from multispectral observed radiances and simulated clear-sky radiances using deep neural networks (DNNs). Two DNN models are established for all-day and daytime-only applications, respectively, using active Cloud Profiling Radar (CPR) and Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) merged cloud products from 2016 as reference labels. The independent dataset from 2017 is used to validate the DNN models. It is shown that both the DNN models outperform the official Moderate Resolution Imaging Spectroradiometer (MODIS) and AHI products in cloud detection and phase discrimination, and the enhancement is more significant over land than over water surface. For multilayer cloud detection, the probability of detecting multilayer clouds reaches ~60% for the all-day model and is increased to ~70% for the daytime model, which is substantially better than MODIS and AHI products. In practical cases, multilayer cloud detection by DNN models is more consistent with CPR/CALIOP than two official products. In addition, the DNN models have superior capability in detecting the optically thin cirrus, which is omitted by MODIS and AHI products. Specifically, the cases also demonstrate that the DNN models can provide effective mixed-phase cloud identification. This deep-learning-based algorithm has the potential for measurements from other similar instruments.

Unsupervised abnormality detection is an appealing approach to identify patterns that are not present in training data without specific annotations for such patterns. In the medical imaging field, methods taking this approach have been proposed to detect lesions. The appeal of this approach stems from the fact that it does not require lesion-specific supervision and can potentially generalize to any sort of abnormal patterns. The principle is to train a generative model on images from healthy individuals to estimate the distribution of images of the normal anatomy, i.e., a normative distribution, and detect lesions as out-of-distribution regions. Restoration-based techniques that modify a given image by taking gradient ascent steps with respect to a posterior distribution composed of a normative distribution and a likelihood term recently yielded state-of-the-art results. However, these methods do not explicitly model ascent directions with respect to the normative distribution, i.e. normative ascent direction, which is essential for successful restoration. In this work, we introduce a novel approach for unsupervised lesion detection by modeling normative ascent directions. We present different modelling options based on the defined ascent directions with local Gaussians. We further extend the proposed method to efficiently utilize 3D information, which has not been explored in most existing works. We experimentally show that the proposed method provides higher accuracy in detection and produces more realistic restored images. The performance of the proposed method is evaluated against baselines on publicly available BRATS and ATLAS stroke lesion datasets; the detection accuracy of the proposed method surpasses the current state-of-the-art results.

Monocular depth estimation has gained increasing attention in recent years, and various techniques have been proposed to tackle this problem. In this work, we aim to provide a comprehensive study on the techniques widely used in monocular depth estimation, and examine their individual influence on the performance. More specifically, we provide a study on: 1) network architectures, including different combinations of encoders/decoders. 2) supervision losses, including fully supervised losses and self-supervised losses and 3) other practices such as input resolution. The experiments are conducted on two commonly used public datasets, KITTI and NYU Depth v2. We also provide an analysis on the errors produced by different models, to reveal the limitations of current methods. Furthermore, by a careful redesign, we present a model for depth estimation, which achieves competitive performance on KITTI and state-of-the-art performance on NYU Depth v2. Our code is publicly available at https://github.com/zenithfang/supervised_dispnet.

As an alternative to manual pixel-wise annotation, synthetic data has been increasingly used for training semantic segmentation models. Such synthetic images and semantic labels can be easily generated from virtual 3D environments. In this work, we propose an approach to cross-domain semantic segmentation with the auxiliary geometric information, which can also be easily obtained from virtual environments. The geometric information is utilized on two levels for reducing domain shift: on the input level, we augment the standard image translation network with the geometric information to translate synthetic images into realistic style; on the output level, we build a task network which simultaneously performs semantic segmentation and depth estimation. Meanwhile, adversarial training is applied on the joint output space to preserve the correlation between semantics and depth. The proposed approach is validated on two pairs of synthetic to real dataset: from Virtual KITTI to KITTI, and from SYNTHIA to Cityscapes, where we achieve a clear performance gain compared to the baselines and various competing methods, demonstrating the effectiveness of the geometric information for cross-domain semantic segmentation.