Open World Compositional Zero-Shot Learning (OW-CZSL) is known to be an extremely challenging task, which aims to recognize unseen compositions formed from seen attributes and objects without any prior assumption of the output space. In order to achieve this goal, a model has to be "smart" and "knowledgeable". To be smart, a model should be good at reasoning the interactions between attributes and objects from the seen compositions. While "knowledgeable" means the model owns "common sense" to the open world that can "foresee" some features of the unseen compositions. Most previous work focuses on the "smart" part, while few of them provided an effective solution to achieve the "knowledgeable" goal. In this paper, we proposed a framework named Multi-Modal Prompt Tuning (MMPT) to inherit the "knowledgeable" property from the large pre-trained vision-language model. Extensive experiments show that our proposed MMPT obtains new state-of-the-art results in OW-CZSL task. On the UT-Zappos dataset, MMPT pushes the AUC score to $29.8$, while the previous best score is $26.5$. On the more challenging MIT-States dataset, the AUC score of MMPT is 1.5 times better than the current state-of-the-art.

To achieve strong real world performance, neural networks must be trained on large, diverse datasets; however, obtaining and annotating such datasets is costly and time-consuming, particularly for 3D point clouds. In this paper, we describe Paved2Paradise, a simple, cost-effective approach for generating fully labeled, diverse, and realistic lidar datasets from scratch, all while requiring minimal human annotation. Our key insight is that, by deliberately collecting separate "background" and "object" datasets (i.e., "factoring the real world"), we can intelligently combine them to produce a combinatorially large and diverse training set. The Paved2Paradise pipeline thus consists of four steps: (1) collecting copious background data, (2) recording individuals from the desired object class(es) performing different behaviors in an isolated environment (like a parking lot), (3) bootstrapping labels for the object dataset, and (4) generating samples by placing objects at arbitrary locations in backgrounds. To demonstrate the utility of Paved2Paradise, we generated synthetic datasets for two tasks: (1) human detection in orchards (a task for which no public data exists) and (2) pedestrian detection in urban environments. Qualitatively, we find that a model trained exclusively on Paved2Paradise synthetic data is highly effective at detecting humans in orchards, including when individuals are heavily occluded by tree branches. Quantitatively, a model trained on Paved2Paradise data that sources backgrounds from KITTI performs comparably to a model trained on the actual dataset. These results suggest the Paved2Paradise synthetic data pipeline can help accelerate point cloud model development in sectors where acquiring lidar datasets has previously been cost-prohibitive.

Understanding videos is an important research topic for multimodal learning. Leveraging large-scale datasets of web-crawled video-text pairs as weak supervision has become a pre-training paradigm for learning joint representations and showcased remarkable potential in video understanding tasks. However, videos can be multi-event and multi-grained, while these video-text pairs usually contain only broad-level video captions. This raises a question: with such weak supervision, can video representation in video-language models gain the ability to distinguish even factual discrepancies in textual description and understand fine-grained events? To address this, we introduce SPOT Prober, to benchmark existing video-language models's capacities of distinguishing event-level discrepancies as an indicator of models' event understanding ability. Our approach involves extracting events as tuples () from videos and generating false event tuples by manipulating tuple components systematically. We reevaluate the existing video-language models with these positive and negative captions and find they fail to distinguish most of the manipulated events. Based on our findings, we propose to plug in these manipulated event captions as hard negative samples and find them effective in enhancing models for event understanding.

Recent advancements in large vision-language models enabled visual object detection in open-vocabulary scenarios, where object classes are defined in free-text formats during inference. In this paper, we aim to probe the state-of-the-art methods for open-vocabulary object detection to determine to what extent they understand fine-grained properties of objects and their parts. To this end, we introduce an evaluation protocol based on dynamic vocabulary generation to test whether models detect, discern, and assign the correct fine-grained description to objects in the presence of hard-negative classes. We contribute with a benchmark suite of increasing difficulty and probing different properties like color, pattern, and material. We further enhance our investigation by evaluating several state-of-the-art open-vocabulary object detectors using the proposed protocol and find that most existing solutions, which shine in standard open-vocabulary benchmarks, struggle to accurately capture and distinguish finer object details. We conclude the paper by highlighting the limitations of current methodologies and exploring promising research directions to overcome the discovered drawbacks. Data and code are available at https://github.com/lorebianchi98/FG-OVD.

In this work, we address the challenging and emergent problem of novel object detection (NOD), focusing on the accurate detection of both known and novel object categories during inference. Traditional object detection algorithms are inherently closed-set, limiting their capability to handle NOD. We present a novel approach to transform existing closed-set detectors into open-set detectors. This transformation is achieved by leveraging the complementary strengths of pre-trained foundational models, specifically CLIP and SAM, through our cooperative mechanism. Furthermore, by integrating this mechanism with state-of-the-art open-set detectors such as GDINO, we establish new benchmarks in object detection performance. Our method achieves 17.42 mAP in novel object detection and 42.08 mAP for known objects on the challenging LVIS dataset. Adapting our approach to the COCO OVD split, we surpass the current state-of-the-art by a margin of 7.2 $ \text{AP}_{50} $ for novel classes. Our code is available at https://github.com/rohit901/cooperative-foundational-models .

Indoor 3D object detection is an essential task in single image scene understanding, impacting spatial cognition fundamentally in visual reasoning. Existing works on 3D object detection from a single image either pursue this goal through independent predictions of each object or implicitly reason over all possible objects, failing to harness relational geometric information between objects. To address this problem, we propose a dynamic sparse graph pipeline named Explicit3D based on object geometry and semantics features. Taking the efficiency into consideration, we further define a relatedness score and design a novel dynamic pruning algorithm followed by a cluster sampling method for sparse scene graph generation and updating. Furthermore, our Explicit3D introduces homogeneous matrices and defines new relative loss and corner loss to model the spatial difference between target pairs explicitly. Instead of using ground-truth labels as direct supervision, our relative and corner loss are derived from the homogeneous transformation, which renders the model to learn the geometric consistency between objects. The experimental results on the SUN RGB-D dataset demonstrate that our Explicit3D achieves better performance balance than the-state-of-the-art.

Loop closure is necessary for correcting errors accumulated in simultaneous localization and mapping (SLAM) in unknown environments. However, conventional loop closure methods based on low-level geometric or image features may cause high ambiguity by not distinguishing similar scenarios. Thus, incorrect loop closures can occur. Though semantic 2D image information is considered in some literature to detect loop closures, there is little work that compares 3D scenes as an integral part of a semantic SLAM system. This paper introduces an approach, called SmSLAM+LCD, integrated into a semantic SLAM system to combine high-level 3D semantic information and low-level feature information to conduct accurate loop closure detection and effective drift reduction. The effectiveness of our approach is demonstrated in testing results.

This paper addresses the problem of enabling a robot to search for a semantic object, i.e., an object with a semantic label, in an unknown and GPS-denied environment. For the robot in the unknown environment to detect and find the target semantic object, it must perform simultaneous localization and mapping (SLAM) at both geometric and semantic levels using its onboard sensors while planning and executing its motion based on the ever-updated SLAM results. In other words, the robot must be able to conduct simultaneous localization, semantic mapping, motion planning, and execution in real-time in the presence of sensing and motion uncertainty. This is an open problem as it combines semantic SLAM based on perception and real-time motion planning and execution under uncertainty. Moreover, the goals of the robot motion change on the fly depending on whether and how the robot can detect the target object. We propose a novel approach to tackle the problem, leveraging semantic SLAM, Bayesian Networks, Markov Decision Process, and Real-Time Dynamic Programming. The results in simulation and real experiments demonstrate the effectiveness and efficiency of our approach.

Accurate perception of objects in the environment is important for improving the scene understanding capability of SLAM systems. In robotic and augmented reality applications, object maps with semantic and metric information show attractive advantages. In this paper, we present RO-MAP, a novel multi-object mapping pipeline that does not rely on 3D priors. Given only monocular input, we use neural radiance fields to represent objects and couple them with a lightweight object SLAM based on multi-view geometry, to simultaneously localize objects and implicitly learn their dense geometry. We create separate implicit models for each detected object and train them dynamically and in parallel as new observations are added. Experiments on synthetic and real-world datasets demonstrate that our method can generate semantic object map with shape reconstruction, and be competitive with offline methods while achieving real-time performance (25Hz). The code and dataset will be available at: https://github.com/XiaoHan-Git/RO-MAP

Semantic 3D mapping, the process of fusing depth and image segmentation information between multiple views to build 3D maps annotated with object classes in real-time, is a recent topic of interest. This paper highlights the fusion overconfidence problem, in which conventional mapping methods assign high confidence to the entire map even when they are incorrect, leading to miscalibrated outputs. Several methods to improve uncertainty calibration at different stages in the fusion pipeline are presented and compared on the ScanNet dataset. We show that the most widely used Bayesian fusion strategy is among the worst calibrated, and propose a learned pipeline that combines fusion and calibration, GLFS, which achieves simultaneously higher accuracy and 3D map calibration while retaining real-time capability. We further illustrate the importance of map calibration on a downstream task by showing that incorporating proper semantic fusion on a modular ObjectNav agent improves its success rates. Our code will be provided on Github for reproducibility upon acceptance.