The automatic generation of stylized co-speech gestures has recently received increasing attention. Previous systems typically allow style control via predefined text labels or example motion clips, which are often not flexible enough to convey user intent accurately. In this work, we present GestureDiffuCLIP, a neural network framework for synthesizing realistic, stylized co-speech gestures with flexible style control. We leverage the power of the large-scale Contrastive-Language-Image-Pre-training (CLIP) model and present a novel CLIP-guided mechanism that extracts efficient style representations from multiple input modalities, such as a piece of text, an example motion clip, or a video. Our system learns a latent diffusion model to generate high-quality gestures and infuses the CLIP representations of style into the generator via an adaptive instance normalization (AdaIN) layer. We further devise a gesture-transcript alignment mechanism that ensures a semantically correct gesture generation based on contrastive learning. Our system can also be extended to allow fine-grained style control of individual body parts. We demonstrate an extensive set of examples showing the flexibility and generalizability of our model to a variety of style descriptions. In a user study, we show that our system outperforms the state-of-the-art approaches regarding human likeness, appropriateness, and style correctness.
This paper considers lossless image compression and presents a learned compression system that can achieve state-of-the-art lossless compression performance but uses only 59K parameters, which is more than 30x less than other learned systems proposed recently in the literature. The explored system is based on a learned pixel-by-pixel lossless image compression method, where each pixel's probability distribution parameters are obtained by processing the pixel's causal neighborhood (i.e. previously encoded/decoded pixels) with a simple neural network comprising 59K parameters. This causality causes the decoder to operate sequentially, i.e. the neural network has to be evaluated for each pixel sequentially, which increases decoding time significantly with common GPU software and hardware. To reduce the decoding time, parallel decoding algorithms are proposed and implemented. The obtained lossless image compression system is compared to traditional and learned systems in the literature in terms of compression performance, encoding-decoding times and computational complexity.
We present a learning framework for reconstructing neural scene representations from a small number of unconstrained tourist photos. Since each image contains transient occluders, decomposing the static and transient components is necessary to construct radiance fields with such in-the-wild photographs where existing methods require a lot of training data. We introduce SF-NeRF, aiming to disentangle those two components with only a few images given, which exploits semantic information without any supervision. The proposed method contains an occlusion filtering module that predicts the transient color and its opacity for each pixel, which enables the NeRF model to solely learn the static scene representation. This filtering module learns the transient phenomena guided by pixel-wise semantic features obtained by a trainable image encoder that can be trained across multiple scenes to learn the prior of transient objects. Furthermore, we present two techniques to prevent ambiguous decomposition and noisy results of the filtering module. We demonstrate that our method outperforms state-of-the-art novel view synthesis methods on Phototourism dataset in a few-shot setting.
Traditional semantic image search methods aim to retrieve images that match the meaning of the text query. However, these methods typically search for objects on the whole image, without considering the localization of objects within the image. This paper presents an extension of existing object detection models for semantic image search that considers the semantic alignment between object proposals and text queries, with a focus on searching for objects within images. The proposed model uses a single feature extractor, a pre-trained Convolutional Neural Network, and a transformer encoder to encode the text query. Proposal-text alignment is performed using contrastive learning, producing a score for each proposal that reflects its semantic alignment with the text query. The Region Proposal Network (RPN) is used to generate object proposals, and the end-to-end training process allows for an efficient and effective solution for semantic image search. The proposed model was trained end-to-end, providing a promising solution for semantic image search that retrieves images that match the meaning of the text query and generates semantically relevant object proposals.
Despite significant efforts, cutting-edge video segmentation methods still remain sensitive to occlusion and rapid movement, due to their reliance on the appearance of objects in the form of object embeddings, which are vulnerable to these disturbances. A common solution is to use optical flow to provide motion information, but essentially it only considers pixel-level motion, which still relies on appearance similarity and hence is often inaccurate under occlusion and fast movement. In this work, we study the instance-level motion and present InstMove, which stands for Instance Motion for Object-centric Video Segmentation. In comparison to pixel-wise motion, InstMove mainly relies on instance-level motion information that is free from image feature embeddings, and features physical interpretations, making it more accurate and robust toward occlusion and fast-moving objects. To better fit in with the video segmentation tasks, InstMove uses instance masks to model the physical presence of an object and learns the dynamic model through a memory network to predict its position and shape in the next frame. With only a few lines of code, InstMove can be integrated into current SOTA methods for three different video segmentation tasks and boost their performance. Specifically, we improve the previous arts by 1.5 AP on OVIS dataset, which features heavy occlusions, and 4.9 AP on YouTubeVIS-Long dataset, which mainly contains fast-moving objects. These results suggest that instance-level motion is robust and accurate, and hence serving as a powerful solution in complex scenarios for object-centric video segmentation.
We introduce an approach to enhance the novel view synthesis from images taken from a freely moving camera. The introduced approach focuses on outdoor scenes where recovering accurate geometric scaffold and camera pose is challenging, leading to inferior results using the state-of-the-art stable view synthesis (SVS) method. SVS and related methods fail for outdoor scenes primarily due to (i) over-relying on the multiview stereo (MVS) for geometric scaffold recovery and (ii) assuming COLMAP computed camera poses as the best possible estimates, despite it being well-studied that MVS 3D reconstruction accuracy is limited to scene disparity and camera-pose accuracy is sensitive to key-point correspondence selection. This work proposes a principled way to enhance novel view synthesis solutions drawing inspiration from the basics of multiple view geometry. By leveraging the complementary behavior of MVS and monocular depth, we arrive at a better scene depth per view for nearby and far points, respectively. Moreover, our approach jointly refines camera poses with image-based rendering via multiple rotation averaging graph optimization. The recovered scene depth and the camera-pose help better view-dependent on-surface feature aggregation of the entire scene. Extensive evaluation of our approach on the popular benchmark dataset, such as Tanks and Temples, shows substantial improvement in view synthesis results compared to the prior art. For instance, our method shows 1.5 dB of PSNR improvement on the Tank and Temples. Similar statistics are observed when tested on other benchmark datasets such as FVS, Mip-NeRF 360, and DTU.
Attention (and distraction) recognition is a key factor in improving human-robot collaboration. We present an assembly scenario where a human operator and a cobot collaborate equally to piece together a gearbox. The setup provides multiple opportunities for the cobot to adapt its behavior depending on the operator's attention, which can improve the collaboration experience and reduce psychological strain. As a first step, we recognize the areas in the workspace that the human operator is paying attention to, and consequently, detect when the operator is distracted. We propose a novel deep-learning approach to develop an attention recognition model. First, we train a convolutional neural network to estimate the gaze direction using a publicly available image dataset. Then, we use transfer learning with a small dataset to map the gaze direction onto pre-defined areas of interest. Models trained using this approach performed very well in leave-one-subject-out evaluation on the small dataset. We performed an additional validation of our models using the video snippets collected from participants working as an operator in the presented assembly scenario. Although the recall for the Distracted class was lower in this case, the models performed well in recognizing the areas the operator paid attention to. To the best of our knowledge, this is the first work that validated an attention recognition model using data from a setting that mimics industrial human-robot collaboration. Our findings highlight the need for validation of attention recognition solutions in such full-fledged, non-guided scenarios.
The diagnostic quality of computed tomography (CT) scans is usually restricted by the induced patient dose, scan speed, and image quality. Sparse-angle tomographic scans reduce radiation exposure and accelerate data acquisition, but suffer from image artifacts and noise. Existing image processing algorithms can restore CT reconstruction quality but often require large training data sets or can not be used for truncated objects. This work presents a self-supervised projection inpainting method that allows learning missing projective views via gradient-based optimization. By reconstructing independent stacks of projection data, a self-supervised loss is calculated in the CT image domain and used to directly optimize projection image intensities to match the missing tomographic views constrained by the projection geometry. Our experiments on real X-ray microscope (XRM) tomographic mouse tibia bone scans show that our method improves reconstructions by 3.1-7.4%/7.7-17.6% in terms of PSNR/SSIM with respect to the interpolation baseline. Our approach is applicable as a flexible self-supervised projection inpainting tool for tomographic applications.
Deep Learning (DL) is one of the most popular research topics in machine learning and DL-driven image recognition systems have developed rapidly. Recent research has used metamorphic testing (MT) to detect misclassified images. Most of them discuss metamorphic relations (MR), with little discussion on which regions should be transformed. We focus on the fact that there are sensitive regions where even a small transformation can easily change the prediction results and propose an MT framework that efficiently tests for regions prone to misclassification by transforming the sensitive regions. Our evaluation showed that the sensitive regions can be specified by Explainable AI (XAI) and our framework effectively detects faults.
The design and evaluation of complex systems can benefit from a software simulation - sometimes called a digital twin. The simulation can be used to characterize system performance or to test its performance under conditions that are difficult to measure (e.g., nighttime for automotive perception systems). We describe the image system simulation software tools that we use to evaluate the performance of image systems for object (automobile) detection. We describe experiments with 13 different cameras with a variety of optics and pixel sizes. To measure the impact of camera spatial resolution, we designed a collection of driving scenes that had cars at many different distances. We quantified system performance by measuring average precision and we report a trend relating system resolution and object detection performance. We also quantified the large performance degradation under nighttime conditions, compared to daytime, for all cameras and a COCO pre-trained network.