35 accepted papers
Large Language Model (LLM) based agents have demonstrated proficiency in multi-step interactions with graphical user interfaces (GUIs). While most research focuses on improving single-task performance, practical scenarios often involve repetitive GUI tasks for which invoking LLM reasoning repeatedly…
Recent work has shown that neural networks can perform 3D tasks such as Novel View Synthesis (NVS) without explicit 3D reconstruction. Even so, we argue that strong 3D inductive biases are still helpful in the design of such networks. We show this point by introducing LagerNVS, an encoder-decoder ne
Bridging the gap between complex human instructions and precise 3D object grounding remains a significant challenge in vision and robotics. Existing 3D segmentation methods often struggle to interpret ambiguous, reasoning-based instructions, while 2D vision-language models that excel at such reasoni
Feed-forward 3D reconstruction from sparse, low-resolution (LR) images is a crucial capability for real-world applications, such as autonomous driving and embodied AI. However, existing methods often fail to recover fine texture details. This limitation stems from the inherent lack of high-frequency
The convergence of 3D geometric perception and video synthesis has created an unprecedented demand for large-scale video data that is rich in both semantic and spatio-temporal information. While existing datasets have advanced either 3D understanding or video generation, a significant gap remains in
Dynamic Gaussian Splatting approaches have achieved remarkable performance for 4D scene reconstruction. However, these approaches rely on dense-frame video sequences for photorealistic reconstruction. In real-world scenarios, due to equipment constraints, sometimes only sparse frames are accessible.
We introduce VGGT-O, a feed-forward model for 3D reconstruction that improves accuracy, efficiency, and capabilities for both static and dynamic scenes. Prior models such as VGGT have shown that feed-forward 3D reconstruction is, in many cases, competitive with traditional optimization-based methods
Long-form video understanding remains challenging due to the extended temporal structure and dense multimodal cues. Despite recent progress, many existing approaches still rely on hand-crafted reasoning pipelines or employ token-consuming video preprocessing to guide MLLMs in autonomous reasoning. T
We introduce WorldGen, a method for generating large, fully formed, navigable 3D worlds from a single text prompt. Existing approaches to 3D scene generation often trade off scene diversity, completeness, and correctness in different ways. We push this envelope by producing large scenes explicitly d
We introduce AutoPartGen, a model that generates objects composed of 3D parts in an autoregressive manner. This model can take as input an image of an object, 2D masks of the object's parts, or an existing 3D object, and generate a corresponding compositional 3D reconstruction. Our approach builds…
Text- or image-to-3D generators and 3D scanners can now produce 3D assets with high-quality shapes and textures, but as single, fused entities lacking meaningful structure. In contrast, most applications and creative workflows require 3D assets to be composed of distinct, meaningful parts that can b…
Accurate trajectory prediction has prominent significance in autonomous driving scenarios. Most existing methods predict the trajectory of an agent by learning its interaction with other agents and the map within the scenario. However, the heterogeneous distribution of these elements across differe…
We present VGGT, a feed-forward neural network that directly infers all key 3D attributes of a scene, including camera parameters, point maps, depth maps, and 3D point tracks, from one, a few, or hundreds of its views. This approach is a step forward in 3D computer vision, where models have typicall…
Benefiting from its hyper-redundant structure, the biomimetic snake-like manipulator retains its remarkable flexibility even within confined spaces. However, its motion planning and control pose significant challenges. This paper imitates the winding uncoiling behavior of snakes to achieve controlla…
Large Language Models (LLM) based agents have shown promise in autonomously completing tasks across various domains, e.g., robotics, games, and web navigation. However, these agents typically require elaborate design and expert prompts to solve tasks in specific domains, which limits their adaptabil…
Predicting future trajectories of traffic agents accurately holds substantial importance in various applications such as autonomous driving. Previous methods commonly infer all future steps of an agent either recursively or simultaneously. However, the recursive strategy suffers from the accumulated…
Recent self-training techniques have shown notable improvements in unsupervised domain adaptation for 3D object detection (3D UDA). These techniques typically select pseudo labels i.e. 3D boxes to supervise models for the target domain. However this selection process inevitably introduces unreliable…
We propose a novel feed-forward 3D editing framework called Shap-Editor. Prior research on editing 3D objects primarily concentrated on editing individual objects by leveraging off-the-shelf 2D image editing networks utilizing a process called 3D distillation which transfers knowledge from the 2D ne…
Contrastive Language-Image Pre-training (CLIP) has demonstrated impressive capabilities in open-vocabulary classification. The class token in the image encoder is trained to capture the global features to distinguish different text descriptions supervised by contrastive loss, making it highly effect…
Operating in narrow spaces is an important challenge in the development of robots. Redundant manipulators are one way to solve this problem, but their mechanism design and control method still have much room for improvement. In this paper, we propose a coiled cable-conduit-driven hyper-redundant man…
In scenarios of the overhead power line system, manual methods are inefficient and unsafe. Meanwhile, the majority of cantilevered robots have poor efficiency when crossing obstacles. This letter proposes a novel power line inspection and maintenance robot to solve these problems. The robot employs
Weakly supervised semantic segmentation (WSSS) with image-level labels is a challenging task. Mainstream approaches follow a multi-stage framework and suffer from high training costs. In this paper, we explore the potential of Contrastive Language-Image Pre-training models (CLIP) to localize differe…
Recent Transformer-based 3D object detectors learn point cloud features either from point- or voxel-based representations. However, the former requires time-consuming sampling while the latter introduces quantization errors. In this paper, we present a novel Point-Voxel Transformer for single-stage…
"Contrastive language-image pretraining has shown great success in learning visual-textual joint representation from web-scale data, demonstrating remarkable “zero-shot” generalization ability for various image tasks. However, how to effectively expand such new language-image pretraining methods to…
Prior works on action representation learning mainly focus on designing various architectures to extract the global representations for short video clips. In contrast, many practical applications such as video alignment have strong demand for learning dense representations for long videos. In this p…
Most neural network pruning methods, such as filter-level and layer-level prunings, prune the network model along one dimension (depth, width, or resolution) solely to meet a computational budget. However, such a pruning policy often leads to excessive reduction of that dimension, thus inducing a hu…
Recently, pure transformer-based models have shown great potentials for vision tasks such as image classification and detection. However, the design of transformer networks is challenging. It has been observed that the depth, embedding dimension, and number of heads can largely affect the performanc…
Federated Learning (FL) as a distributed learning paradigm that aggregates information from diverse clients to train a shared global model, has demonstrated great success. However, malicious clients can perform poisoning attacks and model replacement to introduce backdoors into the trained global mo…
Conventional unsupervised multi-source domain adaptation (UMDA) methods assume all source domains can be accessed directly. However, this assumption neglects the privacy-preserving policy, where all the data and computations must be kept decentralized. There exist three challenges in this scenario:…
Despite remarkable progress achieved, most neural architecture search (NAS) methods focus on searching for one single accurate and robust architecture. To further build models with better generalization capability and performance, model ensemble is usually adopted and performs better than stand-alon…
Relative position encoding (RPE) is important for transformer to capture sequence ordering of input tokens. General efficacy has been proven in natural language processing. However, in computer vision, its efficacy is not well studied and even remains controversial, e.g., whether relative position e…
Semi-supervised variational autoencoders (VAEs) have obtained strong results, but have also encountered the challenge that good ELBO values do not always imply accurate inference results.In this paper, we investigate and propose two causes of this problem: (1) The ELBO objective cannot utilize the l…
Instance segmentation can detect where the objects are in an image, but hard to understand the relationship between them. We pay attention to a typical relationship, relative saliency. A closely related task, salient object detection, predicts a binary map highlighting a visually salient region whil…
Vision Transformer has shown great visual representation power in substantial vision tasks such as recognition and detection, and thus been attracting fast-growing efforts on manually designing more effective architectures. In this paper, we propose to use neural architecture search to automate this…
Deep neural networks for semantic segmentation always require a large number of samples with pixel-level labels, which becomes the major difficulty in their real-world applications. To reduce the labeling cost, unsupervised domain adaptation (UDA) approaches are proposed to transfer knowledge from l…