Posts

Foundational Vision-LLM for AI Linkage and Orchestration

We propose a vision-LLM framework for automating development and deployment of computer vision solutions for pre-defined or custom-defined tasks. A foundational layer is proposed with a code-LLM AI orchestrator self-trained with reinforcement learning to create Python code based on its understanding of a novel user-defined task, together with APIs, documentation and usage notes of existing task-specific AI models. Zero-shot abilities in specific domains are obtained through foundational vision-language models trained at a low compute expense leveraging existing computer vision models and datasets. An engine layer is proposed which comprises of several task-specific vision-language engines which can be compositionally utilized. An application-specific layer is proposed to improve performance in customer-specific scenarios, using novel LLM-guided data augmentation and question decomposition, besides standard fine-tuning tools. We demonstrate a range of applications including visual AI assistance, visual conversation, law enforcement, mobility, medical image reasoning and remote sensing.

Taming Self-Training for Open-Vocabulary Object Detection

Recent studies have shown promising performance in open-vocabulary object detection (OVD) by utilizing pseudo labels (PLs) from pretrained vision and language models (VLMs). However, teacher-student self-training, a powerful and widely used paradigm to leverage PLs, is rarely explored for OVD.

Instantaneous Perception of Moving Objects in 3D

The perception of 3D motion of surrounding traffic participants is crucial for driving safety. While existing works primarily focus on general large motions, we contend that the instantaneous detection and quantification of subtle motions is equally important as they indicate the nuances in driving behavior that may be safety critical, such as behaviors near a stop sign of parking positions. We delve into this under-explored task, examining its unique challenges and developing our solution, accompanied by a carefully designed benchmark. Specifically, due to the lack of correspondences between consecutive frames of sparse Lidar point clouds, static objects might appear to be moving – the so-called swimming effect. This intertwines with the true object motion, thereby posing ambiguity in accurate estimation, especially for subtle motions. To address this, we propose to leverage local occupancy completion of object point clouds to densify the shape cue, and mitigate the impact of swimming artifacts. The occupancy completion is learned in an end-to-end fashion together with the detection of moving objects and the estimation of their motion, instantaneously as soon as objects start to move. Extensive experiments demonstrate superior performance compared to standard 3D motion estimation approaches, particularly highlighting our method’s specialized treatment of subtle motions.

Self-Training Large Language Models for Improved Visual Program Synthesis With Visual Reinforcement

Visual program synthesis is a promising approach to exploit the reasoning abilities of large language models for compositional computer vision tasks. Previous work has used few-shot prompting with frozen LLMs to synthesize visual programs. Training an LLM to write better visual programs is an attractive prospect, but it is unclear how to accomplish this. No dataset of visual programs for training exists, and acquisition of a visual program dataset cannot be easily crowdsourced due to the need for expert annotators. To get around the lack of direct supervision, we explore improving the program synthesis abilities of an LLM using feedback from interactive experience. We propose a method where we exploit existing annotations for a vision-language task to improvise a coarse reward signal for that task, treat the LLM as a policy, and apply reinforced self-training to improve the visual program synthesis ability of the LLM for that task. We describe a series of experiments on object detection, compositional visual question answering, and image-text retrieval, and show that in each case, the self-trained LLM outperforms or performs on par with few-shot frozen LLMs that are an order of magnitude larger. Website: https://zaidkhan.me/ViReP/

AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving

Autonomous vehicle (AV) systems rely on robust perception models as a cornerstone of safety assurance. However, objects encountered on the road exhibit a long-tailed distribution, with rare or unseen categories posing challenges to a deployed perception model. This necessitates an expensive process of continuously curating and annotating data with significant human effort. We propose to leverage recent advances in vision-language and large language models to design an Automatic Data Engine (AIDE) that automatically identifies issues, efficiently curates data, improves the model through auto-labeling, and verifies the model through generation of diverse scenarios. This process operates iteratively, allowing for continuous self-improvement of the model. We further establish a benchmark for open-world detection on AV datasets to comprehensively evaluate various learning paradigms, demonstrating our method’s superior performance at a reduced cost.

LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes

Photorealistic simulation plays a crucial role in applications such as autonomous driving, where advances in neural radiance fields (NeRFs) may allow better scalability through the automatic creation of digital 3D assets. However, reconstruction quality suffers on street scenes due to largely collinear camera motions and sparser samplings at higher speeds. On the other hand, the application often demands rendering from camera views that deviate from the inputs to accurately simulate behaviors like lane changes. In this paper, we propose several insights that allow a better utilization of Lidar data to improve NeRF quality on street scenes. First, our framework learns a geometric scene representation from Lidar, which are fused with the implicit grid-based representation for radiance decoding, thereby supplying strongergeometric information offered by explicit point cloud. Second, we put forth a robust occlusion-aware depth supervision scheme, which allows utilizing densified Lidar points by accumulation. Third, we generate augmented training views from Lidar points for further improvement. Our insights translate to largely improved novel view synthesis under real driving scenes.

Improving the Efficiency-Accuracy Trade-off of DETR-Style Models in Practice

This report aims to provide a comprehensive view on the inference efficiency of DETR-style detection models. We provide the effect of the basic efficiency techniques and identify the factors that are easily applicable yet effectively improve the efficiency-accuracy trade-off. Specifically, we explore the effect of input resolution, multi-scale feature enhancement, and backbone pre-training. Our experiments support that 1) improving the detection accuracy for smaller objects while minimizing the increase in inference cost is a good strategy to achieve a better trade-off between accuracy and efficiency. 2) Multi-scale feature enhancement can be lightened with marginal accuracy loss and 3) improved backbone pre-training can further enhance the trade-off.

Generating Enhanced Negatives for Training Language-Based Object Detectors

The recent progress in language-based open-vocabulary object detection can be largely attributed to finding better ways of leveraging large-scale data with free-form text annotations. Training such models with a discriminative objective function has proven successful, but requires good positive and negative samples.

Long-HOT: A Modular Hierarchical Approach for Long-Horizon Object Transport

We aim to address key challenges in long-horizon embodied exploration and navigation by proposing a long-horizon object transport task called Long-HOT and a novel modular framework for temporally extended navigation. Agents in Long-HOT need to efficiently find and pick up target objects that are scattered in the environment, carry them to a goal location with load constraints, and optionally have access to a container. We propose a modular topological graph-based transport policy (HTP) that explores efficiently with the help of weighted frontiers. Our hierarchical approach uses a combination of motion planning algorithms to reach point goals within explored locations and object navigation policies for moving towards semantic targets at unknown locations. Experiments on both our proposed Habitat transport task and on MultiOn benchmarks show that our method outperforms baselines and prior works. Further, we analyze the agent’s behavior for the usage of the container and demonstrate meaningful generalization to harder transport scenes with training only on simpler versions of the task.

Efficient Transformer Encoders for Mask2Former-style Models

Vision transformer based models bring significant improvements for image segmentation tasks. Although these architectures offer powerful capabilities irrespective of specific segmentation tasks, their use of computational resources can be taxing on deployed devices. One way to overcome this challenge is by adapting the computation level to the specific needs of the input image rather than the current one size-fits-all approach. To this end, we introduce ECO-M2F or EffiCient TransfOrmer Encoders for Mask2Former-style models. Noting that the encoder module of M2F-style models incur high resource-intensive computations, ECO-M2F provides a strategy to self-select the number of hidden layers in the encoder, conditioned on the input image. To enable this self-selection ability for providing a balance between performance and computational efficiency, we present a three-step recipe. The first step is to train the parent architecture to enable early exiting from the encoder. The second step is to create a derived dataset of the ideal number of encoder layers required for each training example. The third step is to use the aforementioned derived dataset to train a gating network that predicts the number of encoder layers to be used, conditioned on input image. Additionally, to change the computational-accuracy trade off, only steps two and three need to be repeated which significantly reduces retraining time. Experiments on the public datasets show that the proposed approach reduces expected encoder computational cost while maintaining performance, adapts to various user compute resources, is flexible in architecture configurations, and can be extended beyond the segmentation task to object detection.