Training a 3D scene understanding model requires complicated human annotations, which are laborious to collect and result in a model only encoding close-set object semantics. In contrast, vision-language pre-training models (e.g., CLIP) have shown remarkable open-world reasoning properties. To this end, we propose directly transferring CLIP's feature space to 3D scene understanding model without any form of supervision. We first modify CLIP's input and forwarding process so that it can be adapted to extract dense pixel features for 3D scene contents. We then project multi-view image features to the point cloud and train a 3D scene understanding model with feature distillation. Without any annotations or additional training, our model achieves promising annotation-free semantic segmentation results on open-vocabulary semantics and long-tailed concepts. Besides, serving as a cross-modal pre-training framework, our method can be used to improve data efficiency during fine-tuning. Our model outperforms previous SOTA methods in various zero-shot and data-efficient learning benchmarks. Most importantly, our model successfully inherits CLIP's rich-structured knowledge, allowing 3D scene understanding models to recognize not only object concepts but also open-world semantics.
Texture analysis is a classical yet challenging task in computer vision for which deep neural networks are actively being applied. Most approaches are based on building feature aggregation modules around a pre-trained backbone and then fine-tuning the new architecture on specific texture recognition tasks. Here we propose a new method named \textbf{R}andom encoding of \textbf{A}ggregated \textbf{D}eep \textbf{A}ctivation \textbf{M}aps (RADAM) which extracts rich texture representations without ever changing the backbone. The technique consists of encoding the output at different depths of a pre-trained deep convolutional network using a Randomized Autoencoder (RAE). The RAE is trained locally to each image using a closed-form solution, and its decoder weights are used to compose a 1-dimensional texture representation that is fed into a linear SVM. This means that no fine-tuning or backpropagation is needed. We explore RADAM on several texture benchmarks and achieve state-of-the-art results with different computational budgets. Our results suggest that pre-trained backbones may not require additional fine-tuning for texture recognition if their learned representations are better encoded.
Modal logics have proved useful for many reasoning tasks in symbolic artificial intelligence (AI), such as belief revision, spatial reasoning, among others. On the other hand, mathematical morphology (MM) is a theory for non-linear analysis of structures, that was widely developed and applied in image analysis. Its mathematical bases rely on algebra, complete lattices, topology. Strong links have been established between MM and mathematical logics, mostly modal logics. In this paper, we propose to further develop and generalize this link between mathematical morphology and modal logic from a topos perspective, i.e. categorial structures generalizing space, and connecting logics, sets and topology. Furthermore, we rely on the internal language and logic of topos. We define structuring elements, dilations and erosions as morphisms. Then we introduce the notion of structuring neighborhoods, and show that the dilations and erosions based on them lead to a constructive modal logic, for which a sound and complete proof system is proposed. We then show that the modal logic thus defined (called morpho-logic here), is well adapted to define concrete and efficient operators for revision, merging, and abduction of new knowledge, or even spatial reasoning.
Text-based Person Search (TPS), is targeted on retrieving pedestrians to match text descriptions instead of query images. Recent Vision-Language Pre-training (VLP) models can bring transferable knowledge to downstream TPS tasks, resulting in more efficient performance gains. However, existing TPS methods improved by VLP only utilize pre-trained visual encoders, neglecting the corresponding textual representation and breaking the significant modality alignment learned from large-scale pre-training. In this paper, we explore the full utilization of textual potential from VLP in TPS tasks. We build on the proposed VLP-TPS baseline model, which is the first TPS model with both pre-trained modalities. We propose the Multi-Integrity Description Constraints (MIDC) to enhance the robustness of the textual modality by incorporating different components of fine-grained corpus during training. Inspired by the prompt approach for zero-shot classification with VLP models, we propose the Dynamic Attribute Prompt (DAP) to provide a unified corpus of fine-grained attributes as language hints for the image modality. Extensive experiments show that our proposed TPS framework achieves state-of-the-art performance, exceeding the previous best method by a margin.
Self-supervised learning has developed rapidly and also advances computer-aided diagnosis in the medical field. Masked image modeling (MIM) is one of the self-supervised learning methods that masks a portion of input pixels and tries to predict the masked pixels. Traditional MIM methods often use a random masking strategy. However, medical images often have a small region of interest for disease detection compared to ordinary images. For example, the regions outside the lung do not contain the information for decision, which may cause the random masking strategy not to learn enough information for COVID-19 detection. Hence, we propose a novel region-guided masked image modeling method (RGMIM) for COVID-19 detection in this paper. In our method, we design a new masking strategy that uses lung mask information to locate valid regions to learn more helpful information for COVID-19 detection. Experimental results show that RGMIM can outperform other state-of-the-art self-supervised learning methods on an open COVID-19 radiography dataset.
AI Generated Content (AIGC) has received tremendous attention within the past few years, with content ranging from image, text, to audio, video, etc. Meanwhile, AIGC has become a double-edged sword and recently received much criticism regarding its responsible usage. In this vision paper, we focus on three main concerns that may hinder the healthy development and deployment of AIGC in practice, including risks from privacy, bias, toxicity, misinformation, and intellectual property (IP). By documenting known and potential risks, as well as any possible misuse scenarios of AIGC, the aim is to draw attention to potential risks and misuse, help society to eliminate obstacles, and promote the more ethical and secure deployment of AIGC. Additionally, we provide insights into the promising directions for tackling these risks while constructing generative models, enabling AIGC to be used responsibly to benefit society.
The image-based diagnosis is now a vital aspect of modern automation assisted diagnosis. To enable models to produce pixel-level diagnosis, pixel-level ground-truth labels are essentially required. However, since it is often not straight forward to obtain the labels in many application domains such as in medical image, classification-based approaches have become the de facto standard to perform the diagnosis. Though they can identify class-salient regions, they may not be useful for diagnosis where capturing all of the evidences is important requirement. Alternatively, a counterfactual explanation (CX) aims at providing explanations using a casual reasoning process of form "If X has not happend, Y would not heppend". Existing CX approaches, however, use classifier to explain features that can change its predictions. Thus, they can only explain class-salient features, rather than entire object of interest. This hence motivates us to propose a novel CX strategy that is not reliant on image classification. This work is inspired from the recent developments in generative adversarial networks (GANs) based image-to-image domain translation, and leverages to translate an abnormal image to counterpart normal image (i.e. counterfactual instance CI) to find discrepancy maps between the two. Since it is generally not possible to obtain abnormal and normal image pairs, we leverage Cycle-Consistency principle (a.k.a CycleGAN) to perform the translation in unsupervised way. We formulate CX in terms of a discrepancy map that, when added from the abnormal image, will make it indistinguishable from the CI. We evaluate our method on three datasets including a synthetic, tuberculosis and BraTS dataset. All these experiments confirm the supremacy of propose method in generating accurate CX and CI.
Recent advances in robot learning have shown promise in enabling robots to perform a variety of manipulation tasks and generalize to novel scenarios. One of the key contributing factors to this progress is the scale of robot data used to train the models. To obtain large-scale datasets, prior approaches have relied on either demonstrations requiring high human involvement or engineering-heavy autonomous data collection schemes, both of which are challenging to scale. To mitigate this issue, we propose an alternative route and leverage text-to-image foundation models widely used in computer vision and natural language processing to obtain meaningful data for robot learning without requiring additional robot data. We term our method Robot Learning with Semantically Imagened Experience (ROSIE). Specifically, we make use of the state of the art text-to-image diffusion models and perform aggressive data augmentation on top of our existing robotic manipulation datasets via inpainting various unseen objects for manipulation, backgrounds, and distractors with text guidance. Through extensive real-world experiments, we show that manipulation policies trained on data augmented this way are able to solve completely unseen tasks with new objects and can behave more robustly w.r.t. novel distractors. In addition, we find that we can improve the robustness and generalization of high-level robot learning tasks such as success detection through training with the diffusion-based data augmentation. The project's website and videos can be found at diffusion-rosie.github.io
Hyperspectral image is unique and useful for its abundant spectral bands, but it subsequently requires extra elaborated treatments of the spatial-spectral correlation as well as the global correlation along the spectrum for building a robust and powerful HSI restoration algorithm. By considering such HSI characteristics, 3D Quasi-Recurrent Neural Network (QRNN3D) is one of the HSI denoising networks that has been shown to achieve excellent performance and flexibility. In this paper, we show that with a few simple modifications, the performance of QRNN3D could be substantially improved further. Our modifications are based on the finding that through QRNN3D is powerful for modeling spectral correlation, it neglects the proper treatment between features from different sources and its training strategy is suboptimal. We, therefore, introduce an adaptive fusion module to replace its vanilla additive skip connection to better fuse the features of the encoder and decoder. We additionally identify several important techniques to further enhance the performance, which includes removing batch normalization, use of extra frequency loss, and learning rate warm-up. Experimental results on various noise settings demonstrate the effectiveness and superior performance of our method.
Traffic scene parsing is one of the most important tasks to achieve intelligent cities. So far, little effort has been spent on constructing datasets specifically for the task of traffic scene parsing. To fill this gap, here we introduce the TSP6K dataset, containing 6,000 urban traffic images and spanning hundreds of street scenes under various weather conditions. In contrast to most previous traffic scene datasets collected from a driving platform, the images in our dataset are from the shooting platform high-hanging on the street. Such traffic images can capture more crowded street scenes with several times more traffic participants than the driving scenes. Each image in the TSP6K dataset is provided with high-quality pixel-level and instance-level annotations. We perform a detailed analysis for the dataset and comprehensively evaluate the state-of-the-art scene parsing methods. Considering the vast difference in instance sizes, we propose a detail refining decoder, which recovers the details of different semantic regions in traffic scenes. Experiments have shown its effectiveness in parsing high-hanging traffic scenes. Code and dataset will be made publicly available.