Search

The scaling of large language models to encode all the world's knowledge in model parameters is unsustainable and has exacerbated resource barriers. Retrieval-Augmented Generation (RAG) presents a potential solution, yet its application to vision-language models (VLMs) is under explored. Existing methods focus on models designed for single tasks. Furthermore, they're limited by the need for resource intensive pre training, additional parameter requirements, unaddressed modality prioritization and lack of clear benefit over non-retrieval baselines. This paper introduces RAVEN, a multitask retrieval augmented VLM framework that enhances base VLMs through efficient, task specific fine-tuning. By integrating retrieval augmented samples without the need for additional retrieval-specific parameters, we show that the model acquires retrieval properties that are effective across multiple tasks. Our results and extensive ablations across retrieved modalities for the image captioning and VQA tasks indicate significant performance improvements compared to non retrieved baselines +1 CIDEr on MSCOCO, +4 CIDEr on NoCaps and nearly a +3\% accuracy on specific VQA question types. This underscores the efficacy of applying RAG approaches to VLMs, marking a stride toward more efficient and accessible multimodal learning.

Central Pattern Generators (CPGs) form the neural basis of the observed rhythmic behaviors for locomotion in legged animals. The CPG dynamics organized into networks allow the emergence of complex locomotor behaviors. In this work, we take this inspiration for developing walking behaviors in multi-legged robots. We present novel DeepCPG policies that embed CPGs as a layer in a larger neural network and facilitate end-to-end learning of locomotion behaviors in deep reinforcement learning (DRL) setup. We demonstrate the effectiveness of this approach on physics engine-based insectoid robots. We show that, compared to traditional approaches, DeepCPG policies allow sample-efficient end-to-end learning of effective locomotion strategies even in the case of high-dimensional sensor spaces (vision). We scale the DeepCPG policies using a modular robot configuration and multi-agent DRL. Our results suggest that gradual complexification with embedded priors of these policies in a modular fashion could achieve non-trivial sensor and motor integration on a robot platform. These results also indicate the efficacy of bootstrapping more complex intelligent systems from simpler ones based on biological principles. Finally, we present the experimental results for a proof-of-concept insectoid robot system for which DeepCPG learned policies initially using the simulation engine and these were afterwards transferred to real-world robots without any additional fine-tuning., Comment: Preprint of paper accepted for publication in IEEE Transaction On Cognitive and Developmental Systems

We demonstrate the classification of common motions of held objects using the harmonic micro-Doppler signatures scattered from harmonic radio-frequency tags. Harmonic tags capture incident signals and retransmit at harmonic frequencies, making them easier to distinguish from clutter. We characterize the motion of tagged handheld objects via the time-varying frequency shift of the harmonic signals (harmonic Doppler). With complex micromotions of held objects, the time-frequency response manifests complex micro-Doppler signatures that can be used to classify the motions. We developed narrow-band harmonic tags at 2.4/4.8 GHz that support frequency scalability for multi-tag operation, and a harmonic radar system to transmit a 2.4 GHz continuous-wave signal and receive the scattered 4.8 GHz harmonic signal. Experiments were conducted to mimic four common motions of held objects from 35 subjects in a cluttered indoor environment. A 7-layer convolutional neural network (CNN) multi-label classifier was developed and obtained a real time classification accuracy of 94.24%, with a response time of 2 seconds per sample with a data processing latency of less than 0.5 seconds., Comment: Pages: 7 Figures: 9 Tables: 1

Deep reinforcement learning (RL) has made it possible to solve complex robotics problems using neural networks as function approximators. However, the policies trained on stationary environments suffer in terms of generalization when transferred from one environment to another. In this work, we use Robust Markov Decision Processes (RMDP) to train the drone control policy, which combines ideas from Robust Control and RL. It opts for pessimistic optimization to handle potential gaps between policy transfer from one environment to another. The trained control policy is tested on the task of quadcopter positional control. RL agents were trained in a MuJoCo simulator. During testing, different environment parameters (unseen during the training) were used to validate the robustness of the trained policy for transfer from one environment to another. The robust policy outperformed the standard agents in these environments, suggesting that the added robustness increases generality and can adapt to non-stationary environments. Codes: https://github.com/adipandas/gym_multirotor, Comment: 6 pages; 3 Figures; Accepted in https://mecc2021.a2c2.org/

Fine-tuning from a collection of models pre-trained on different domains (a "model zoo") is emerging as a technique to improve test accuracy in the low-data regime. However, model selection, i.e. how to pre-select the right model to fine-tune from a model zoo without performing any training, remains an open topic. We use a linearized framework to approximate fine-tuning, and introduce two new baselines for model selection -- Label-Gradient and Label-Feature Correlation. Since all model selection algorithms in the literature have been tested on different use-cases and never compared directly, we introduce a new comprehensive benchmark for model selection comprising of: i) A model zoo of single and multi-domain models, and ii) Many target tasks. Our benchmark highlights accuracy gain with model zoo compared to fine-tuning Imagenet models. We show our model selection baseline can select optimal models to fine-tune in few selections and has the highest ranking correlation to fine-tuning accuracy compared to existing algorithms., Comment: 14 pages

In this paper, we present a novel developmental reinforcement learning-based controller for a quadcopter with thrust vectoring capabilities. This multirotor UAV design has tilt-enabled rotors. It utilizes the rotor force magnitude and direction to achieve the desired state during flight. The control policy of this robot is learned using the policy transfer from the learned controller of the quadcopter (comparatively simple UAV design without thrust vectoring). This approach allows learning a control policy for systems with multiple inputs and multiple outputs. The performance of the learned policy is evaluated by physics-based simulations for the tasks of hovering and way-point navigation. The flight simulations utilize a flight controller based on reinforcement learning without any additional PID components. The results show faster learning with the presented approach as opposed to learning the control policy from scratch for this new UAV design created by modifications in a conventional quadcopter, i.e., the addition of more degrees of freedom (4-actuators in conventional quadcopter to 8-actuators in tilt-rotor quadcopter). We demonstrate the robustness of our learned policy by showing the recovery of the tilt-rotor platform in the simulation from various non-static initial conditions in order to reach a desired state. The developmental policy for the tilt-rotor UAV also showed superior fault tolerance when compared with the policy learned from the scratch. The results show the ability of the presented approach to bootstrap the learned behavior from a simpler system (lower-dimensional action-space) to a more complex robot (comparatively higher-dimensional action-space) and reach better performance faster., Comment: 10 pages, 8 figures, Accepted in Dynamic Systems and Control Conference (https://event.asme.org/DSCC)

In this paper, we present an autonomous flight controller for a quadcopter with thrust vectoring capabilities. This UAV falls in the category of multirotors with tilt-motion enabled rotors. Since the vehicle considered is over-actuated in nature, the dynamics and control allocation have to be analysed carefully. Moreover, the possibility of hovering at large attitude maneuvers of this novel vehicle requires singularity-free attitude control. Hence, quaternion state feedback is utilized to compute the control commands for the UAV motors while avoiding the gimbal lock condition experienced by Euler angle based controllers. The quaternion implementation also reduces the overall complexity of state estimation due to absence of trigonometric parameters. The quadcopter dynamic model and state space is utilized to design the attitude controller and control allocation for the UAV. The control allocation, in particular, is derived by linearizing the system about hover condition. This mathematical method renders the control allocation more accurate than existing approaches. Lyapunov stability analysis of the attitude controller is shown to prove global stability. The quaternion feedback attitude controller is commanded by an outer position controller loop which generates rotor-tilt and desired quaternions commands for the system. The performance of the UAV is evaluated by numerical simulations for tracking attitude step commands and for following a way-point navigation mission., Comment: Accepted for publication in American Controls Conference 2020, 6-Pages, 10 figures

Quality control is an essential process in manufacturing to make the product defect-free as well as to meet customer needs. The automation of this process is important to maintain high quality along with the high manufacturing throughput. With recent developments in deep learning and computer vision technologies, it has become possible to detect various features from the images with near-human accuracy. However, many of these approaches are data intensive. Training and deployment of such a system on manufacturing floors may become expensive and time-consuming. The need for large amounts of training data is one of the limitations of the applicability of these approaches in real-world manufacturing systems. In this work, we propose the application of a Siamese convolutional neural network to do one-shot recognition for such a task. Our results demonstrate how one-shot learning can be used in quality control of steel by identification of defects on the steel surface. This method can significantly reduce the requirements of training data and can also be run in real-time., Comment: Accepted for publication in NAMRC 48

Digitization has led to smart, connected technologies be an integral part of businesses, governments and communities. For manufacturing digitization, there has been active research and development with a focus on Cloud Manufacturing (CM) and the Industrial Internet of Things (IIoT). This work presents a computer vision toolkit (CV Toolkit) for non-invasive digitization of the factory floor in line with Industry 4.0 requirements for factory data collection. Currently, technical challenges persist towards digitization of legacy systems due to the limitation for changes in their design and sensors. This novel toolkit is developed to facilitate easy integration of legacy production machinery and factory floor artifacts with the digital and smart manufacturing environment with no requirement of any physical changes in the machines. The system developed is modular, and allows real-time monitoring of production machinery. Modularity aspect allows the incorporation of new software applications in the current framework of CV Toolkit. To allow connectivity of this toolkit with manufacturing floors in a simple, deployable and cost-effective manner, the toolkit is integrated with a known manufacturing data standard, MTConnect, to "translate" the digital inputs into data streams that can be read by commercial status tracking and reporting software solutions. The proposed toolkit is demonstrated using a mock-panel environment developed in house at the University of Cincinnati to highlight its usability., Comment: Accepted for publication in 48th SME North American Manufacturing Research Conference (NAMRC48)

The conceptual design and flight controller of a novel kind of quadcopter are presented. This design is capable of morphing the shape of the UAV during flight to achieve position and attitude control. We consider a dynamic center of gravity (CoG) which causes continuous variation in a moment of inertia (MoI) parameters of the UAV in this design. These dynamic structural parameters play a vital role in the stability and control of the system. The length of quadcopter arms is a variable parameter, and it is actuated using attitude feedback-based control law. The MoI parameters are computed in real-time and incorporated in the equations of motion of the system. The UAV utilizes the angular motion of propellers and variable quadcopter arm lengths for position and navigation control. The movement space of the CoG is a design parameter and it is bounded by actuator limitations and stability requirements of the system. A detailed information on equations of motion, flight controller design and possible applications of this system are provided. Further, the proposed shape-changing UAV system is evaluated by comparative numerical simulations for way point navigation mission and complex trajectory tracking., Comment: 6 Pages

Image captioning is an ambiguous problem, with many suitable captions for an image. To address ambiguity, beam search is the de facto method for sampling multiple captions. However, beam search is computationally expensive and known to produce generic captions. To address this concern, some variational auto-encoder (VAE) and generative adversarial net (GAN) based methods have been proposed. Though diverse, GAN and VAE are less accurate. In this paper, we first predict a meaningful summary of the image, then generate the caption based on that summary. We use part-of-speech as summaries, since our summary should drive caption generation. We achieve the trifecta: (1) High accuracy for the diverse captions as evaluated by standard captioning metrics and user studies; (2) Faster computation of diverse captions compared to beam search and diverse beam search; and (3) High diversity as evaluated by counting novel sentences, distinct n-grams and mutual overlap (i.e., mBleu-4) scores., Comment: 12 pages with references and appendix. To appear CVPR'19

Image captioning is an important but challenging task, applicable to virtual assistants, editing tools, image indexing, and support of the disabled. Its challenges are due to the variability and ambiguity of possible image descriptions. In recent years significant progress has been made in image captioning, using Recurrent Neural Networks powered by long-short-term-memory (LSTM) units. Despite mitigating the vanishing gradient problem, and despite their compelling ability to memorize dependencies, LSTM units are complex and inherently sequential across time. To address this issue, recent work has shown benefits of convolutional networks for machine translation and conditional image generation. Inspired by their success, in this paper, we develop a convolutional image captioning technique. We demonstrate its efficacy on the challenging MSCOCO dataset and demonstrate performance on par with the baseline, while having a faster training time per number of parameters. We also perform a detailed analysis, providing compelling reasons in favor of convolutional language generation approaches., Comment: 11 pages, 9 Figures

Colorization is an ambiguous problem, with multiple viable colorizations for a single grey-level image. However, previous methods only produce the single most probable colorization. Our goal is to model the diversity intrinsic to the problem of colorization and produce multiple colorizations that display long-scale spatial co-ordination. We learn a low dimensional embedding of color fields using a variational autoencoder (VAE). We construct loss terms for the VAE decoder that avoid blurry outputs and take into account the uneven distribution of pixel colors. Finally, we build a conditional model for the multi-modal distribution between grey-level image and the color field embeddings. Samples from this conditional model result in diverse colorization. We demonstrate that our method obtains better diverse colorizations than a standard conditional variational autoencoder (CVAE) model, as well as a recently proposed conditional generative adversarial network (cGAN)., Comment: This revision to appear in CVPR17

We show how to extend traditional intrinsic image decompositions to incorporate further layers above albedo and shading. It is hard to obtain data to learn a multi-layer decomposition. Instead, we can learn to decompose an image into layers that are "like this" by authoring generative models for each layer using proxy examples that capture the Platonic ideal (Mondrian images for albedo; rendered 3D primitives for shading; material swatches for shading detail). Our method then generates image layers, one from each model, that explain the image. Our approach rests on innovation in generative models for images. We introduce a Convolutional Variational Auto Encoder (conv-VAE), a novel VAE architecture that can reconstruct high fidelity images. The approach is general, and does not require that layers admit a physical interpretation.

Problems such as predicting a new shading field (Y) for an image (X) are ambiguous: many very distinct solutions are good. Representing this ambiguity requires building a conditional model P(Y|X) of the prediction, conditioned on the image. Such a model is difficult to train, because we do not usually have training data containing many different shadings for the same image. As a result, we need different training examples to share data to produce good models. This presents a danger we call "code space collapse" - the training procedure produces a model that has a very good loss score, but which represents the conditional distribution poorly. We demonstrate an improved method for building conditional models by exploiting a metric constraint on training data that prevents code space collapse. We demonstrate our model on two example tasks using real data: image saturation adjustment, image relighting. We describe quantitative metrics to evaluate ambiguous generation results. Our results quantitatively and qualitatively outperform different strong baselines.

Several methods have been proposed for large-scale 3D reconstruction from large, unorganized image collections. A large reconstruction problem is typically divided into multiple components which are reconstructed independently using structure from motion (SFM) and later merged together. Incremental SFM methods are most popular for the basic structure recovery of a single component. They are robust and effective but are strictly sequential in nature. We present a multistage approach for SFM reconstruction of a single component that breaks the sequential nature of the incremental SFM methods. Our approach begins with quickly building a coarse 3D model using only a fraction of features from given images. The coarse model is then enriched by localizing remaining images and matching and triangulating remaining features in subsequent stages. These stages are made efficient and highly parallel by leveraging the geometry of the coarse model. Our method produces similar quality models as compared to incremental SFM methods while being notably fast and parallel.

Detecting multiple planes in images is a challenging problem, but one with many applications. Recent work such as J-Linkage and Ordered Residual Kernels have focussed on developing a domain independent approach to detect multiple structures. These multiple structure detection methods are then used for estimating multiple homographies given feature matches between two images. Features participating in the multiple homographies detected, provide us the multiple scene planes. We show that these methods provide locally optimal results and fail to merge detected planar patches to the true scene planes. These methods use only residues obtained on applying homography of one plane to another as cue for merging. In this paper, we develop additional cues such as local consistency of planes, local normals, texture etc. to perform better classification and merging . We formulate the classification as an MRF problem and use TRWS message passing algorithm to solve non metric energy terms and complex sparse graph structure. We show results on challenging dataset common in robotics navigation scenarios where our method shows accuracy of more than 85 percent on average while being close or same as the actual number of scene planes., Comment: 6 pages, 22 figures, ICPR conference

Parallel computing using accelerators has gained widespread research attention in the past few years. In particular, using GPUs for general purpose computing has brought forth several success stories with respect to time taken, cost, power, and other metrics. However, accelerator based computing has signifi- cantly relegated the role of CPUs in computation. As CPUs evolve and also offer matching computational resources, it is important to also include CPUs in the computation. We call this the hybrid computing model. Indeed, most computer systems of the present age offer a degree of heterogeneity and therefore such a model is quite natural. We reevaluate the claim of a recent paper by Lee et al.(ISCA 2010). We argue that the right question arising out of Lee et al. (ISCA 2010) should be how to use a CPU+GPU platform efficiently, instead of whether one should use a CPU or a GPU exclusively. To this end, we experiment with a set of 13 diverse workloads ranging from databases, image processing, sparse matrix kernels, and graphs. We experiment with two different hybrid platforms: one consisting of a 6-core Intel i7-980X CPU and an NVidia Tesla T10 GPU, and another consisting of an Intel E7400 dual core CPU with an NVidia GT520 GPU. On both these platforms, we show that hybrid solutions offer good advantage over CPU or GPU alone solutions. On both these platforms, we also show that our solutions are 90% resource efficient on average. Our work therefore suggests that hybrid computing can offer tremendous advantages at not only research-scale platforms but also the more realistic scale systems with significant performance gains and resource efficiency to the large scale user community., Comment: 20 pages