Recently, Bird's-Eye-View (BEV) representation has gained increasing attention in multi-view 3D object detection, which has demonstrated promising applications in autonomous driving. Although multi-view camera systems can be deployed at low cost, the lack of depth information makes current approaches adopt large models for good performance. Therefore, it is essential to improve the efficiency of BEV 3D object detection. Knowledge Distillation (KD) is one of the most practical techniques to train efficient yet accurate models. However, BEV KD is still under-explored to the best of our knowledge. Different from image classification tasks, BEV 3D object detection approaches are more complicated and consist of several components. In this paper, we propose a unified framework named BEV-LGKD to transfer the knowledge in the teacher-student manner. However, directly applying the teacher-student paradigm to BEV features fails to achieve satisfying results due to heavy background information in RGB cameras. To solve this problem, we propose to leverage the localization advantage of LiDAR points. Specifically, we transform the LiDAR points to BEV space and generate the foreground mask and view-dependent mask for the teacher-student paradigm. It is to be noted that our method only uses LiDAR points to guide the KD between RGB models. As the quality of depth estimation is crucial for BEV perception, we further introduce depth distillation to our framework. Our unified framework is simple yet effective and achieves a significant performance boost. Code will be released.
Data diversity and volume are crucial to the success of training deep learning models, while in the medical imaging field, the difficulty and cost of data collection and annotation are especially huge. Specifically in robotic surgery, data scarcity and imbalance have heavily affected the model accuracy and limited the design and deployment of deep learning-based surgical applications such as surgical instrument segmentation. Considering this, in this paper, we rethink the surgical instrument segmentation task and propose a one-to-many data generation solution that gets rid of the complicated and expensive process of data collection and annotation from robotic surgery. In our method, we only utilize a single surgical background tissue image and a few open-source instrument images as the seed images and apply multiple augmentations and blending techniques to synthesize amounts of image variations. In addition, we also introduce the chained augmentation mixing during training to further enhance the data diversities. The proposed approach is evaluated on the real datasets of the EndoVis-2018 and EndoVis-2017 surgical scene segmentation. Our empirical analysis suggests that without the high cost of data collection and annotation, we can achieve decent surgical instrument segmentation performance. Moreover, we also observe that our method can deal with novel instrument prediction in the deployment domain. We hope our inspiring results would encourage researchers to emphasize data-centric methods to overcome demanding deep learning limitations besides data shortage, such as class imbalance, domain adaptation, and incremental learning.
We present, NIO - Neural Interpolation Operator, a lightweight efficient neural operator-based architecture to perform video frame interpolation. Current deep learning based methods rely on local convolutions for feature learning and require a large amount of training on comprehensive datasets. Furthermore, transformer-based architectures are large and need dedicated GPUs for training. On the other hand, NIO, our neural operator-based approach learns the features in the frames by translating the image matrix into the Fourier space by using Fast Fourier Transform (FFT). The model performs global convolution, making it discretization invariant. We show that NIO can produce visually-smooth and accurate results and converges in fewer epochs than state-of-the-art approaches. To evaluate the visual quality of our interpolated frames, we calculate the structural similarity index (SSIM) and Peak Signal to Noise Ratio (PSNR) between the generated frame and the ground truth frame. We provide the quantitative performance of our model on Vimeo-90K dataset, DAVIS, UCF101 and DISFA+ dataset.
Self-supervised pre-training techniques have achieved remarkable progress in Document AI. Most multimodal pre-trained models use a masked language modeling objective to learn bidirectional representations on the text modality, but they differ in pre-training objectives for the image modality. This discrepancy adds difficulty to multimodal representation learning. In this paper, we propose LayoutLMv3 to pre-train multimodal Transformers for Document AI with unified text and image masking. Additionally, LayoutLMv3 is pre-trained with a word-patch alignment objective to learn cross-modal alignment by predicting whether the corresponding image patch of a text word is masked. The simple unified architecture and training objectives make LayoutLMv3 a general-purpose pre-trained model for both text-centric and image-centric Document AI tasks. Experimental results show that LayoutLMv3 achieves state-of-the-art performance not only in text-centric tasks, including form understanding, receipt understanding, and document visual question answering, but also in image-centric tasks such as document image classification and document layout analysis. The code and models are publicly available at https://aka.ms/layoutlmv3.
Endoscopy is the most widely used imaging technique for the diagnosis of cancerous lesions in hollow organs. However, endoscopic images are often affected by illumination artefacts: image parts may be over- or underexposed according to the light source pose and the tissue orientation. These artifacts have a strong negative impact on the performance of computer vision or AI-based diagnosis tools. Although endoscopic image enhancement methods are greatly required, little effort has been devoted to over- and under-exposition enhancement in real-time. This contribution presents an extension to the objective function of LMSPEC, a method originally introduced to enhance images from natural scenes. It is used here for the exposure correction in endoscopic imaging and the preservation of structural information. To the best of our knowledge, this contribution is the first one that addresses the enhancement of endoscopic images using deep learning (DL) methods. Tested on the Endo4IE dataset, the proposed implementation has yielded a significant improvement over LMSPEC reaching a SSIM increase of 4.40% and 4.21% for over- and underexposed images, respectively.
We present a method to edit complex indoor lighting from a single image with its predicted depth and light source segmentation masks. This is an extremely challenging problem that requires modeling complex light transport, and disentangling HDR lighting from material and geometry with only a partial LDR observation of the scene. We tackle this problem using two novel components: 1) a holistic scene reconstruction method that estimates scene reflectance and parametric 3D lighting, and 2) a neural rendering framework that re-renders the scene from our predictions. We use physically-based indoor light representations that allow for intuitive editing, and infer both visible and invisible light sources. Our neural rendering framework combines physically-based direct illumination and shadow rendering with deep networks to approximate global illumination. It can capture challenging lighting effects, such as soft shadows, directional lighting, specular materials, and interreflections. Previous single image inverse rendering methods usually entangle scene lighting and geometry and only support applications like object insertion. Instead, by combining parametric 3D lighting estimation with neural scene rendering, we demonstrate the first automatic method to achieve full scene relighting, including light source insertion, removal, and replacement, from a single image. All source code and data will be publicly released.
Despite recent success in conditional image synthesis, prevalent input conditions such as semantics and edges are not clear enough to express `Linear (Ridges)' and `Planar (Scale)' representations. To address this problem, we propose a novel framework StyLandGAN, which synthesizes desired landscape images using a depth map which has higher expressive power. Our StyleLandGAN is extended from the unconditional generation model to accept input conditions. We also propose a '2-phase inference' pipeline which generates diverse depth maps and shifts local parts so that it can easily reflect user's intend. As a comparison, we modified the existing semantic image synthesis models to accept a depth map as well. Experimental results show that our method is superior to existing methods in quality, diversity, and depth-accuracy.
Most interpretability research focuses on datasets containing thousands of images of commonplace objects. However, many high-impact datasets, such as those in medicine and the geosciences, contain fine-grain objects that require domain-expert knowledge to recognize and are time-consuming to collect and annotate. As a result, these datasets contain few annotated images, and current machine vision models cannot train intensively on them. Thus, adapting interpretability techniques to maximize the amount of information that models can learn from small, fine-grain datasets is an important endeavor. Using a Mask R-CNN to segment ancient reef fossils in rock sample images, we present a general paradigm for identifying and mitigating model weaknesses. Specifically, we apply image perturbations to expose the Mask R-CNN's inability to distinguish between different classes of fossils and its inconsistency in segmenting fossils with different textures. To address these shortcomings, we extend an existing model-editing method for correcting systematic mistakes in image classification to image segmentation and introduce a novel application of the technique: encouraging a greater separation between positive and negative pixels for a given class. Through extensive experiments, we find that editing the model by perturbing all pixels for a given class in one image is most effective (compared to using multiple images and/or fewer pixels). Our paradigm may also generalize to other segmentation models trained on small, fine-grain datasets.
State estimation in complex illumination environments based on conventional visual-inertial odometry is a challenging task due to the severe visual degradation of the visual camera. The thermal infrared camera is capable of all-day time and is less affected by illumination variation. However, most existing visual data association algorithms are incompatible because the thermal infrared data contains large noise and low contrast. Motivated by the phenomenon that thermal radiation varies most significantly at the edges of objects, the study proposes an ETIO, which is the first edge-based monocular thermal-inertial odometry for robust localization in visually degraded environments. Instead of the raw image, we utilize the binarized image from edge extraction for pose estimation to overcome the poor thermal infrared image quality. Then, an adaptive feature tracking strategy ADT-KLT is developed for robust data association based on limited edge information and its distance distribution. Finally, a pose graph optimization performs real-time estimation over a sliding window of recent states by combining IMU pre-integration with reprojection error of all edge feature observations. We evaluated the performance of the proposed system on public datasets and real-world experiments and compared it against state-of-the-art methods. The proposed ETIO was verified with the ability to enable accurate and robust localization all-day time.
PICNet pioneered the generation of multiple and diverse results for image completion task, but it required a careful balance between $\mathcal{KL}$ loss (diversity) and reconstruction loss (quality), resulting in a limited diversity and quality . Separately, iGPT-based architecture has been employed to infer distributions in a discrete space derived from a pixel-level pre-clustered palette, which however cannot generate high-quality results directly. In this work, we present a novel framework for pluralistic image completion that can achieve both high quality and diversity at much faster inference speed. The core of our design lies in a simple yet effective code sharing mechanism that leads to a very compact yet expressive image representation in a discrete latent domain. The compactness and the richness of the representation further facilitate the subsequent deployment of a transformer to effectively learn how to composite and complete a masked image at the discrete code domain. Based on the global context well-captured by the transformer and the available visual regions, we are able to sample all tokens simultaneously, which is completely different from the prevailing autoregressive approach of iGPT-based works, and leads to more than 100$\times$ faster inference speed. Experiments show that our framework is able to learn semantically-rich discrete codes efficiently and robustly, resulting in much better image reconstruction quality. Our diverse image completion framework significantly outperforms the state-of-the-art both quantitatively and qualitatively on multiple benchmark datasets.