In this work, we construct the largest dataset for multimodal pretraining in Chinese, which consists of over 1.9TB images and 292GB texts that cover a wide range of domains. We propose a cross-modal pretraining method called M6, referring to Multi-Modality to Multi-Modality Multitask Mega-transformer, for unified pretraining on the data of single modality and multiple modalities. We scale the model size up to 10 billion and 100 billion parameters, and build the largest pretrained model in Chinese. We apply the model to a series of downstream applications, and demonstrate its outstanding performance in comparison with strong baselines. Furthermore, we specifically design a downstream task of text-guided image generation, and show that the finetuned M6 can create high-quality images with high resolution and abundant details.
With the rapid deployment of service robots, a method should be established to allow multiple robots to work in the same place to collaborate and share the spatial information. To this end, we present a collaborative visual simultaneous localization and mapping (SLAM) framework particularly designed for service robot scenarios. With an edge server maintaining a map database and performing global optimization, each robot can register to an existing map, update the map, or build new maps, all with a unified interface and low computation and memory cost. To enable real-time information sharing, an efficient landmark retrieval method is proposed to allow each robot to get nearby landmarks observed by others. The framework is general enough to support both RGB-D and monocular cameras, as well as robots with multiple cameras, taking the rigid constraints between cameras into consideration. The proposed framework has been fully implemented and verified with public datasets and live experiments.
Holistically understanding an object and its 3D movable parts through visual perception models is essential for enabling an autonomous agent to interact with the world. For autonomous driving, the dynamics and states of vehicle parts such as doors, the trunk, and the bonnet can provide meaningful semantic information and interaction states, which are essential to ensuring the safety of the self-driving vehicle. Existing visual perception models mainly focus on coarse parsing such as object bounding box detection or pose estimation and rarely tackle these situations. In this paper, we address this important autonomous driving problem by solving three critical issues. First, to deal with data scarcity, we propose an effective training data generation process by fitting a 3D car model with dynamic parts to vehicles in real images before reconstructing human-vehicle interaction (VHI) scenarios. Our approach is fully automatic without any human interaction, which can generate a large number of vehicles in uncommon states (VUS) for training deep neural networks (DNNs). Second, to perform fine-grained vehicle perception, we present a multi-task network for VUS parsing and a multi-stream network for VHI parsing. Third, to quantitatively evaluate the effectiveness of our data augmentation approach, we build the first VUS dataset in real traffic scenarios (e.g., getting on/out or placing/removing luggage). Experimental results show that our approach advances other baseline methods in 2D detection and instance segmentation by a big margin (over 8%). In addition, our network yields large improvements in discovering and understanding these uncommon cases. Moreover, we have released the source code, the dataset, and the trained model on Github (https://github.com/zongdai/EditingForDNN).
Using radiological scans to identify liver tumors is crucial for proper patient treatment. This is highly challenging, as top radiologists only achieve F1 scores of roughly 80% (hepatocellular carcinoma (HCC) vs. others) with only moderate inter-rater agreement, even when using multi-phase magnetic resonance (MR) imagery. Thus, there is great impetus for computer-aided diagnosis (CAD) solutions. A critical challengeis to reliably parse a 3D MR volume to localize diagnosable regions of interest (ROI). In this paper, we break down this problem using a key-slice parser (KSP), which emulates physician workflows by first identifying key slices and then localize their corresponding key ROIs. Because performance demands are so extreme, (not to miss any key ROI),our KSP integrates complementary modules--top-down classification-plus-detection (CPD) and bottom-up localization-by-over-segmentation(LBOS). The CPD uses a curve-parsing and detection confidence to re-weight classifier confidences. The LBOS uses over-segmentation to flag CPD failure cases and provides its own ROIs. For scalability, LBOS is only weakly trained on pseudo-masks using a new distance-aware Tversky loss. We evaluate our approach on the largest multi-phase MR liver lesion test dataset to date (430 biopsy-confirmed patients). Experiments demonstrate that our KSP can localize diagnosable ROIs with high reliability (85% patients have an average overlap of >= 40% with the ground truth). Moreover, we achieve an HCC vs. others F1 score of 0.804, providing a fully-automated CAD solution comparable with top human physicians.
Natural language processing (NLP) has been applied to various fields including text classification and sentiment analysis. In the shared task of sentiment analysis of code-mixed tweets, which is a part of the SemEval-2020 competition~\cite{patwa2020sentimix}, we preprocess datasets by replacing emoji and deleting uncommon characters and so on, and then fine-tune the Bidirectional Encoder Representation from Transformers(BERT) to perform the best. After exhausting top3 submissions, Our team MeisterMorxrc achieves an averaged F1 score of 0.730 in this task, and and our codalab username is MeisterMorxrc.
Texts appearing in daily scenes that can be recognized by OCR (Optical Character Recognition) tools contain significant information, such as street name, product brand and prices. Two tasks -- text-based visual question answering and text-based image captioning, with a text extension from existing vision-language applications, are catching on rapidly. To address these problems, many sophisticated multi-modality encoding frameworks (such as heterogeneous graph structure) are being used. In this paper, we argue that a simple attention mechanism can do the same or even better job without any bells and whistles. Under this mechanism, we simply split OCR token features into separate visual- and linguistic-attention branches, and send them to a popular Transformer decoder to generate answers or captions. Surprisingly, we find this simple baseline model is rather strong -- it consistently outperforms state-of-the-art (SOTA) models on two popular benchmarks, TextVQA and all three tasks of ST-VQA, although these SOTA models use far more complex encoding mechanisms. Transferring it to text-based image captioning, we also surpass the TextCaps Challenge 2020 winner. We wish this work to set the new baseline for this two OCR text related applications and to inspire new thinking of multi-modality encoder design. Code is available at https://github.com/ZephyrZhuQi/ssbaseline
To achieve robust motion estimation in visually degraded environments, thermal odometry has been an attraction in the robotics community. However, most thermal odometry methods are purely based on classical feature extractors, which is difficult to establish robust correspondences in successive frames due to sudden photometric changes and large thermal noise. To solve this problem, we propose ThermalPoint, a lightweight feature detection network specifically tailored for producing keypoints on thermal images, providing notable anti-noise improvements compared with other state-of-the-art methods. After that, we combine ThermalPoint with a novel radiometric feature tracking method, which directly makes use of full radiometric data and establishes reliable correspondences between sequential frames. Finally, taking advantage of an optimization-based visual-inertial framework, a deep feature-based thermal-inertial odometry (TP-TIO) framework is proposed and evaluated thoroughly in various visually degraded environments. Experiments show that our method outperforms state-of-the-art visual and laser odometry methods in smoke-filled environments and achieves competitive accuracy in normal environments.
Recently, hyperspectral image (HSI) classification approaches based on deep learning (DL) models have been proposed and shown promising performance. However, because of very limited available training samples and massive model parameters, DL methods may suffer from overfitting. In this paper, we propose an end-to-end 3-D lightweight convolutional neural network (CNN) (abbreviated as 3-D-LWNet) for limited samples-based HSI classification. Compared with conventional 3-D-CNN models, the proposed 3-D-LWNet has a deeper network structure, less parameters, and lower computation cost, resulting in better classification performance. To further alleviate the small sample problem, we also propose two transfer learning strategies: 1) cross-sensor strategy, in which we pretrain a 3-D model in the source HSI data sets containing a greater number of labeled samples and then transfer it to the target HSI data sets and 2) cross-modal strategy, in which we pretrain a 3-D model in the 2-D RGB image data sets containing a large number of samples and then transfer it to the target HSI data sets. In contrast to previous approaches, we do not impose restrictions over the source data sets, in which they do not have to be collected by the same sensors as the target data sets. Experiments on three public HSI data sets captured by different sensors demonstrate that our model achieves competitive performance for HSI classification compared to several state-of-the-art methods