Hung-Ting Su

Kuan-Chih Huang, Tsung-Han Wu, Hung-Ting Su et al.

Monocular 3D object detection is an important yet challenging task in autonomous driving. Some existing methods leverage depth information from an off-the-shelf depth estimator to assist 3D detection, but suffer from the additional computational burden and achieve limited performance caused by inaccurate depth priors. To alleviate this, we propose MonoDTR, a novel end-to-end depth-aware transformer network for monocular 3D object detection. It mainly consists of two components: (1) the Depth-Aware Feature Enhancement (DFE) module that implicitly learns depth-aware features with auxiliary supervision without requiring extra computation, and (2) the Depth-Aware Transformer (DTR) module that globally integrates context- and depth-aware features. Moreover, different from conventional pixel-wise positional encodings, we introduce a novel depth positional encoding (DPE) to inject depth positional hints into transformers. Our proposed depth-aware modules can be easily plugged into existing image-only monocular 3D object detectors to improve the performance. Extensive experiments on the KITTI dataset demonstrate that our approach outperforms previous state-of-the-art monocular-based methods and achieves real-time detection. Code is available at https://github.com/kuanchihhuang/MonoDTR

Gueter Josmy Faure, Min-Hung Chen, Jia-Fong Yeh et al.

Vision-Language Models (VLMs) have demonstrated remarkable capabilities in general video understanding, yet they often struggle with the fine-grained comprehension crucial for real-world applications requiring nuanced interpretation of human actions and interactions. While some recent human-centric benchmarks evaluate aspects of model behaviour such as fairness/ethics, emotion perception, and broader human-centric metrics, they do not combine long-form videos, very dense QA coverage, and frame-level spatial/temporal grounding at scale. To bridge this gap, we introduce FineBench, a human-centric video question answering (VQA) benchmark specifically designed to assess fine-grained understanding. FineBench comprises 199,420 multiple-choice QA pairs densely annotated across 64 long-form videos (15 minutes each), focusing on detailed person movement, person interaction, and object manipulation, including compositional actions. Our extensive evaluation reveals that while proprietary models like GPT-5 achieve respectable performance, current open-source VLMs significantly underperform, struggling particularly with spatial reasoning in multi-person scenes and distinguishing subtle differences in human movements and interactions. To address these identified weaknesses, we propose FineAgent, a modular framework that enhances VLMs by leveraging a Localizer and a Descriptor. Experiments show that FineAgent consistently improves the performance of various open VLMs on FineBench. FineBench provides a rigorous testbed for future research into fine-grained human-centric video understanding, while FineAgent offers a practical approach to enhance such reasoning in current VLMs.

Hung-Ting Su, Yulei Niu, Xudong Lin et al.

Causal Video Question Answering (CVidQA) queries not only association or temporal relations but also causal relations in a video. Existing question synthesis methods pre-trained question generation (QG) systems on reading comprehension datasets with text descriptions as inputs. However, QG models only learn to ask association questions (e.g., ``what is someone doing...'') and result in inferior performance due to the poor transfer of association knowledge to CVidQA, which focuses on causal questions like ``why is someone doing ...''. Observing this, we proposed to exploit causal knowledge to generate question-answer pairs, and proposed a novel framework, Causal Knowledge Extraction from Language Models (CaKE-LM), leveraging causal commonsense knowledge from language models to tackle CVidQA. To extract knowledge from LMs, CaKE-LM generates causal questions containing two events with one triggering another (e.g., ``score a goal'' triggers ``soccer player kicking ball'') by prompting LM with the action (soccer player kicking ball) to retrieve the intention (to score a goal). CaKE-LM significantly outperforms conventional methods by 4% to 6% of zero-shot CVidQA accuracy on NExT-QA and Causal-VidQA datasets. We also conduct comprehensive analyses and provide key findings for future research.

Chien Cheng Chyou, Hung-Ting Su, Winston H. Hsu

Adversarial robustness poses a critical challenge in the deployment of deep learning models for real-world applications. Traditional approaches to adversarial training and supervised detection rely on prior knowledge of attack types and access to labeled training data, which is often impractical. Existing unsupervised adversarial detection methods identify whether the target model works properly, but they suffer from bad accuracies owing to the use of common cross-entropy training loss, which relies on unnecessary features and strengthens adversarial attacks. We propose new training losses to reduce useless features and the corresponding detection method without prior knowledge of adversarial attacks. The detection rate (true positive rate) against all given white-box attacks is above 93.9% except for attacks without limits (DF($\infty$)), while the false positive rate is barely 2.5%. The proposed method works well in all tested attack types and the false positive rates are even better than the methods good at certain types.

Hsin-Ying Lee, Hung-Ting Su, Bing-Chen Tsai et al.

While recent large-scale video-language pre-training made great progress in video question answering, the design of spatial modeling of video-language models is less fine-grained than that of image-language models; existing practices of temporal modeling also suffer from weak and noisy alignment between modalities. To learn fine-grained visual understanding, we decouple spatial-temporal modeling and propose a hybrid pipeline, Decoupled Spatial-Temporal Encoders, integrating an image- and a video-language encoder. The former encodes spatial semantics from larger but sparsely sampled frames independently of time, while the latter models temporal dynamics at lower spatial but higher temporal resolution. To help the video-language model learn temporal relations for video QA, we propose a novel pre-training objective, Temporal Referring Modeling, which requires the model to identify temporal positions of events in video sequences. Extensive experiments demonstrate that our model outperforms previous work pre-trained on orders of magnitude larger datasets.

Tsung-Han Wu, Hung-Ting Su, Shang-Tse Chen et al.

Fairness and robustness play vital roles in trustworthy machine learning. Observing safety-critical needs in various annotation-expensive vision applications, we introduce a novel learning framework, Fair Robust Active Learning (FRAL), generalizing conventional active learning to fair and adversarial robust scenarios. This framework allows us to achieve standard and robust minimax fairness with limited acquired labels. In FRAL, we then observe existing fairness-aware data selection strategies suffer from either ineffectiveness under severe data imbalance or inefficiency due to huge computations of adversarial training. To address these two problems, we develop a novel Joint INconsistency (JIN) method exploiting prediction inconsistencies between benign and adversarial inputs as well as between standard and robust models. These two inconsistencies can be used to identify potential fairness gains and data imbalance mitigations. Thus, by performing label acquisition with our inconsistency-based ranking metrics, we can alleviate the class imbalance issue and enhance minimax fairness with limited computation. Extensive experiments on diverse datasets and sensitive groups demonstrate that our method obtains the best results in standard and robust fairness under white-box PGD attacks compared with existing active data selection baselines.

Gueter Josmy Faure, Jia-Fong Yeh, Min-Hung Chen et al.

Long-form video understanding presents unique challenges that extend beyond traditional short-video analysis approaches, particularly in capturing long-range dependencies, processing redundant information efficiently, and extracting high-level semantic concepts. To address these challenges, we propose a novel approach that more accurately reflects human cognition. This paper introduces HERMES: temporal-coHERent long-forM understanding with Episodes and Semantics, featuring two versatile modules that can enhance existing video-language models or operate as a standalone system. Our Episodic COmpressor (ECO) efficiently aggregates representations from micro to semi-macro levels, reducing computational overhead while preserving temporal dependencies. Our Semantics ReTRiever (SeTR) enriches these representations with semantic information by focusing on broader context, dramatically reducing feature dimensionality while preserving relevant macro-level information. We demonstrate that these modules can be seamlessly integrated into existing SOTA models, consistently improving their performance while reducing inference latency by up to 43% and memory usage by 46%. As a standalone system, HERMES achieves state-of-the-art performance across multiple long-video understanding benchmarks in both zero-shot and fully-supervised settings.

Hung-Ting Su, Ting-Jun Wang, Jia-Fong Yeh et al.

Conventional Vision-and-Language Navigation (VLN) benchmarks assume instructions are feasible and the referenced target exists, leaving agents ill-equipped to handle false-premise goals. We introduce VLN-NF, a benchmark with false-premise instructions where the target is absent from the specified room and agents must navigate, gather evidence through in-room exploration, and explicitly output NOT-FOUND. VLN-NF is constructed via a scalable pipeline that rewrites VLN instructions using an LLM and verifies target absence with a VLM, producing plausible yet factually incorrect goals. We further propose REV-SPL to jointly evaluate room reaching, exploration coverage, and decision correctness. To address this challenge, we present ROAM, a two-stage hybrid that combines supervised room-level navigation with LLM/VLM-driven in-room exploration guided by a free-space clearance prior. ROAM achieves the best REV-SPL among compared methods, while baselines often under-explore and terminate prematurely under unreliable instructions. VLN-NF project page can be found at https://vln-nf.github.io/.

Hung-Ting Su, Ya-Ching Hsu, Xudong Lin et al.

Large language models (LLMs) equipped with chain-of-thoughts (CoT) prompting have shown significant multi-step reasoning capabilities in factual content like mathematics, commonsense, and logic. However, their performance in narrative reasoning, which demands greater abstraction capabilities, remains unexplored. This study utilizes tropes in movie synopses to assess the abstract reasoning abilities of state-of-the-art LLMs and uncovers their low performance. We introduce a trope-wise querying approach to address these challenges and boost the F1 score by 11.8 points. Moreover, while prior studies suggest that CoT enhances multi-step reasoning, this study shows CoT can cause hallucinations in narrative content, reducing GPT-4's performance. We also introduce an Adversarial Injection method to embed trope-related text tokens into movie synopses without explicit tropes, revealing CoT's heightened sensitivity to such injections. Our comprehensive analysis provides insights for future research directions.

Pei-An Chen, Yong-Ching Liang, Jia-Fong Yeh et al.

Intelligent embodied agents should not simply follow instructions, as real-world environments often involve unexpected conditions and exceptions. However, existing methods usually focus on directly executing instructions, without considering whether the target objects can actually be manipulated, meaning they fail to assess available affordances. To address this limitation, we introduce DynAfford, a benchmark that evaluates embodied agents in dynamic environments where object affordances may change over time and are not specified in the instruction. DynAfford requires agents to perceive object states, infer implicit preconditions, and adapt their actions accordingly. To enable this capability, we introduce ADAPT, a plug-and-play module that augments existing planners with explicit affordance reasoning. Experiments demonstrate that incorporating ADAPT significantly improves robustness and task success across both seen and unseen environments. We also show that a domain-adapted, LoRA-finetuned vision-language model used as the affordance inference backend outperforms a commercial LLM (GPT-4o), highlighting the importance of task-aligned affordance grounding.

Tsung-Han Wu, Hung-Ting Su, Shang-Tse Chen et al.

The robust self-training (RST) framework has emerged as a prominent approach for semi-supervised adversarial training. To explore the possibility of tackling more complicated tasks with even lower labeling budgets, unlike prior approaches that rely on robust pretrained models, we present SNORD - a simple yet effective framework that introduces contemporary semi-supervised learning techniques into the realm of adversarial training. By enhancing pseudo labels and managing noisy training data more effectively, SNORD showcases impressive, state-of-the-art performance across diverse datasets and labeling budgets, all without the need for pretrained models. Compared to full adversarial supervision, SNORD achieves a 90% relative robust accuracy under epsilon = 8/255 AutoAttack, requiring less than 0.1%, 2%, and 10% labels for CIFAR-10, CIFAR-100, and TinyImageNet-200, respectively. Additional experiments confirm the efficacy of each component and demonstrate the adaptability of integrating SNORD with existing adversarial pretraining strategies to further bolster robustness.

Pawat Chunhachatrachai, Gueter Josmy Faure, Hung-Ting Su et al.

Spatial question answering over egocentric video is a challenging task that requires Vision-Language Models (VLMs) to reason about 3D object positions, scene affordances, and directional relationships, particularly in the zero-shot setting where no task-specific fine-tuning is available. We introduce SpatioRoute, a dynamic prompt generation approach that routes each incoming question to a semantically tailored prompt template -- without any additional training, fine-tuning, or 3D sensor input. SpatioRoute operates in two complementary modes: SpatioRoute-R, a rule-based router that deterministically maps question typologies (e.g., What, Is, How, Can, Which) to specialized prompt templates; and SpatioRoute-L, an LLM-driven approach that generates task-specific prompts from the question and situational context alone, with no video input at routing time. We evaluate SpatioRoute on the SQA3D benchmark across VLMs spanning model families. SpatioRoute achieves consistent overall accuracy gains up to 5% over fixed prompt baselines, establishing a new state-of-the-art for zero-shot video-only spatial VQA without requiring 3D point-cloud inputs. As an additional finding, we observe that Chain-of-Thought (CoT) prompting, implemented via the Think it Twice architecture, consistently degrades performance in this setting on Qwen series models, confirming that question-aware routing is more effective than uniform reasoning instructions for spatial video understanding.

Tzu-Jung Lin, Jia-Fong Yeh, Hung-Ting Su et al.

In open-vocabulary mobile manipulation (OVMM), task success often hinges on the selection of an appropriate base placement for the robot. Existing approaches typically navigate to proximity-based regions without considering affordances, resulting in frequent manipulation failures. We propose Affordance-Guided Coarse-to-Fine Exploration, a zero-shot framework for base placement that integrates semantic understanding from vision-language models (VLMs) with geometric feasibility through an iterative optimization process. Our method constructs cross-modal representations, namely Affordance RGB and Obstacle Map+, to align semantics with spatial context. This enables reasoning that extends beyond the egocentric limitations of RGB perception. To ensure interaction is guided by task-relevant affordances, we leverage coarse semantic priors from VLMs to guide the search toward task-relevant regions and refine placements with geometric constraints, thereby reducing the risk of convergence to local optima. Evaluated on five diverse open-vocabulary mobile manipulation tasks, our system achieves an 85% success rate, significantly outperforming classical geometric planners and VLM-based methods. This demonstrates the promise of affordance-aware and multimodal reasoning for generalizable, instruction-conditioned planning in OVMM.

Posheng Chen, Powen Cheng, Gueter Josmy Faure et al.

In real-world scenes, target objects may reside in regions that are not visible. While humans can often infer the locations of occluded objects from context and commonsense knowledge, this capability remains a major challenge for vision-language models (VLMs). To address this gap, we introduce SceneFunRI, a benchmark for Reasoning the Invisible. Based on the SceneFun3D dataset, SceneFunRI formulates the task as a 2D spatial reasoning problem via a semi-automatic pipeline and comprises 855 instances. It requires models to infer the locations of invisible functional objects from task instructions and commonsense reasoning. The strongest baseline model (Gemini 3 Flash) only achieves an CAcc@75 of 15.20, an mIoU of 0.74, and a Dist of 28.65. We group our prompting analysis into three categories: Strong Instruction Prompting, Reasoning-based Prompting, and Spatial Process of Elimination (SPoE). These findings indicate that invisible-region reasoning remains an unstable capability in current VLMs, motivating future work on models that more tightly integrate task intent, commonsense priors, spatial grounding, and uncertainty-aware search.

Tsung-Han Wu, Yueh-Cheng Liu, Yu-Kai Huang et al.

Despite the success of deep learning on supervised point cloud semantic segmentation, obtaining large-scale point-by-point manual annotations is still a significant challenge. To reduce the huge annotation burden, we propose a Region-based and Diversity-aware Active Learning (ReDAL), a general framework for many deep learning approaches, aiming to automatically select only informative and diverse sub-scene regions for label acquisition. Observing that only a small portion of annotated regions are sufficient for 3D scene understanding with deep learning, we use softmax entropy, color discontinuity, and structural complexity to measure the information of sub-scene regions. A diversity-aware selection algorithm is also developed to avoid redundant annotations resulting from selecting informative but similar regions in a querying batch. Extensive experiments show that our method highly outperforms previous active learning strategies, and we achieve the performance of 90% fully supervised learning, while less than 15% and 5% annotations are required on S3DIS and SemanticKITTI datasets, respectively. Our code is publicly available at https://github.com/tsunghan-wu/ReDAL.

Ke-Jyun Wang, Yun-Hsuan Liu, Hung-Ting Su et al.

To effectively apply robots in working environments and assist humans, it is essential to develop and evaluate how visual grounding (VG) can affect machine performance on occluded objects. However, current VG works are limited in working environments, such as offices and warehouses, where objects are usually occluded due to space utilization issues. In our work, we propose a novel OCID-Ref dataset featuring a referring expression segmentation task with referring expressions of occluded objects. OCID-Ref consists of 305,694 referring expressions from 2,300 scenes with providing RGB image and point cloud inputs. To resolve challenging occlusion issues, we argue that it's crucial to take advantage of both 2D and 3D signals to resolve challenging occlusion issues. Our experimental results demonstrate the effectiveness of aggregating 2D and 3D signals but referring to occluded objects still remains challenging for the modern visual grounding systems. OCID-Ref is publicly available at https://github.com/lluma/OCID-Ref

ChungYi Lin, Shen-Lung Tung, Hung-Ting Su et al.

To address the limitations of traffic prediction from location-bound detectors, we present Geographical Cellular Traffic (GCT) flow, a novel data source that leverages the extensive coverage of cellular traffic to capture mobility patterns. Our extensive analysis validates its potential for transportation. Focusing on vehicle-related GCT flow prediction, we propose a graph neural network that integrates multivariate, temporal, and spatial facets for improved accuracy. Experiments reveal our model's superiority over baselines, especially in long-term predictions. We also highlight the potential for GCT flow integration into transportation systems.

Ting-Kang Yen, Igor Morawski, Shusil Dangi et al.

Event-based object detection has recently garnered attention in the computer vision community due to the exceptional properties of event cameras, such as high dynamic range and no motion blur. However, feature asynchronism and sparsity cause invisible objects due to no relative motion to the camera, posing a significant challenge in the task. Prior works have studied various implicit-learned memories to retain as many temporal cues as possible. However, implicit memories still struggle to preserve long-term features effectively. In this paper, we consider those invisible objects as pseudo-occluded objects and aim to detect them by tracking through occlusions. Firstly, we introduce the visibility attribute of objects and contribute an auto-labeling algorithm to not only clean the existing event camera dataset but also append additional visibility labels to it. Secondly, we exploit tracking strategies for pseudo-occluded objects to maintain their permanence and retain their bounding boxes, even when features have not been available for a very long time. These strategies can be treated as an explicit-learned memory guided by the tracking objective to record the displacements of objects across frames. Lastly, we propose a spatio-temporal feature aggregation module to enrich the latent features and a consistency loss to increase the robustness of the overall pipeline. We conduct comprehensive experiments to verify our method's effectiveness where still objects are retained, but real occluded objects are discarded. The results demonstrate that (1) the additional visibility labels can assist in supervised training, and (2) our method outperforms state-of-the-art approaches with a significant improvement of 7.9% absolute mAP.

Gueter Josmy Faure, Min-Hung Chen, Jia-Fong Yeh et al.

This paper introduces MovieCORE, a novel video question answering (VQA) dataset designed to probe deeper cognitive understanding of movie content. Unlike existing datasets that focus on surface-level comprehension, MovieCORE emphasizes questions that engage System-2 thinking while remaining specific to the video material. We present an innovative agentic brainstorming approach, utilizing multiple large language models (LLMs) as thought agents to generate and refine high-quality question-answer pairs. To evaluate dataset quality, we develop a set of cognitive tests assessing depth, thought-provocation potential, and syntactic complexity. We also propose a comprehensive evaluation scheme for assessing VQA model performance on deeper cognitive tasks. To address the limitations of existing video-language models (VLMs), we introduce an agentic enhancement module, Agentic Choice Enhancement (ACE), which improves model reasoning capabilities post-training by up to 25%. Our work contributes to advancing movie understanding in AI systems and provides valuable insights into the capabilities and limitations of current VQA models when faced with more challenging, nuanced questions about cinematic content. Our project page, dataset and code can be found at https://joslefaure.github.io/assets/html/moviecore.html.

Hung-Ting Su, Chun-Tong Chao, Ya-Ching Hsu et al.

Large Language Models (LLMs) have demonstrated effectiveness not only in language tasks but also in video reasoning. This paper introduces a novel dataset, Tropes in Movies (TiM), designed as a testbed for exploring two critical yet previously overlooked video reasoning skills: (1) Abstract Perception: understanding and tokenizing abstract concepts in videos, and (2) Long-range Compositional Reasoning: planning and integrating intermediate reasoning steps for understanding long-range videos with numerous frames. Utilizing tropes from movie storytelling, TiM evaluates the reasoning capabilities of state-of-the-art LLM-based approaches. Our experiments show that current methods, including Captioner-Reasoner, Large Multimodal Model Instruction Fine-tuning, and Visual Programming, only marginally outperform a random baseline when tackling the challenges of Abstract Perception and Long-range Compositional Reasoning. To address these deficiencies, we propose Face-Enhanced Viper of Role Interactions (FEVoRI) and Context Query Reduction (ConQueR), which enhance Visual Programming by fostering role interaction awareness and progressively refining movie contexts and trope queries during reasoning processes, significantly improving performance by 15 F1 points. However, this performance still lags behind human levels (40 vs. 65 F1). Additionally, we introduce a new protocol to evaluate the necessity of Abstract Perception and Long-range Compositional Reasoning for task resolution. This is done by analyzing the code generated through Visual Programming using an Abstract Syntax Tree (AST), thereby confirming the increased complexity of TiM. The dataset and code are available at: https://ander1119.github.io/TiM

ChungYi Lin, Shen-Lung Tung, Hung-Ting Su et al.

Vehicle flow, a crucial indicator for transportation, is often limited by detector coverage. With the advent of extensive mobile network coverage, we can leverage mobile user activities, or cellular traffic, on roadways as a proxy for vehicle flow. However, as counts of cellular traffic may not directly align with vehicle flow due to data from various user types, we present a new task: predicting vehicle flow in camera-free areas using cellular traffic. To uncover correlations within multi-source data, we deployed cameras on selected roadways to establish the Tel2Veh dataset, consisting of extensive cellular traffic and sparse vehicle flows. Addressing this challenge, we propose a framework that independently extracts features and integrates them with a graph neural network (GNN)-based fusion to discern disparities, thereby enabling the prediction of unseen vehicle flows using cellular traffic. This work advances the use of telecom data in transportation and pioneers the fusion of telecom and vision-based data, offering solutions for traffic management.

Jia-Fong Yeh, Chi-Ming Chung, Hung-Ting Su et al.

In few-shot imitation learning (FSIL), using behavioral cloning (BC) to solve unseen tasks with few expert demonstrations becomes a popular research direction. The following capabilities are essential in robotics applications: (1) Behaving in compound tasks that contain multiple stages. (2) Retrieving knowledge from few length-variant and misalignment demonstrations. (3) Learning from a different expert. No previous work can achieve these abilities at the same time. In this work, we conduct FSIL problem under the union of above settings and introduce a novel stage conscious attention network (SCAN) to retrieve knowledge from few demonstrations simultaneously. SCAN uses an attention module to identify each stage in length-variant demonstrations. Moreover, it is designed under demonstration-conditioned policy that learns the relationship between experts and agents. Experiment results show that SCAN can learn from different experts without fine-tuning and outperform baselines in complicated compound tasks with explainable visualization.

Guan-Rong Lu, Yueh-Cheng Liu, Tung-I Chen et al.

Anomaly awareness is an essential capability for safety-critical applications such as autonomous driving. While recent progress of robotics and computer vision has enabled anomaly detection for image classification, anomaly detection on semantic segmentation is less explored. Conventional anomaly-aware systems assuming other existing classes as out-of-distribution (pseudo-unknown) classes for training a model will result in two drawbacks. (1) Unknown classes, which applications need to cope with, might not actually exist during training time. (2) Model performance would strongly rely on the class selection. Observing this, we propose a novel Synthetic-Unknown Data Generation, intending to tackle the anomaly-aware semantic segmentation task. We design a new Masked Gradient Update (MGU) module to generate auxiliary data along the boundary of in-distribution data points. In addition, we modify the traditional cross-entropy loss to emphasize the border data points. We reach the state-of-the-art performance on two anomaly segmentation datasets. Ablation studies also demonstrate the effectiveness of proposed modules.

Chung-Yi Lin, Hung-Ting Su, Shen-Lung Tung et al.

Spatial-temporal prediction is a critical problem for intelligent transportation, which is helpful for tasks such as traffic control and accident prevention. Previous studies rely on large-scale traffic data collected from sensors. However, it is unlikely to deploy sensors in all regions due to the device and maintenance costs. This paper addresses the problem via outdoor cellular traffic distilled from over two billion records per day in a telecom company, because outdoor cellular traffic induced by user mobility is highly related to transportation traffic. We study road intersections in urban and aim to predict future outdoor cellular traffic of all intersections given historic outdoor cellular traffic. Furthermore, we propose a new model for multivariate spatial-temporal prediction, mainly consisting of two extending graph attention networks (GAT). First GAT is used to explore correlations among multivariate cellular traffic. Another GAT leverages the attention mechanism into graph propagation to increase the efficiency of capturing spatial dependency. Experiments show that the proposed model significantly outperforms the state-of-the-art methods on our dataset.

Hung-Ting Su, Po-Wei Shen, Bing-Chen Tsai et al.

Understanding and comprehending video content is crucial for many real-world applications such as search and recommendation systems. While recent progress of deep learning has boosted performance on various tasks using visual cues, deep cognition to reason intentions, motivation, or causality remains challenging. Existing datasets that aim to examine video reasoning capability focus on visual signals such as actions, objects, relations, or could be answered utilizing text bias. Observing this, we propose a novel task, along with a new dataset: Trope Understanding in Movies and Animations (TrUMAn), with 2423 videos associated with 132 tropes, intending to evaluate and develop learning systems beyond visual signals. Tropes are frequently used storytelling devices for creative works. By coping with the trope understanding task and enabling the deep cognition skills of machines, data mining applications and algorithms could be taken to the next level. To tackle the challenging TrUMAn dataset, we present a Trope Understanding and Storytelling (TrUSt) with a new Conceptual Storyteller module, which guides the video encoder by performing video storytelling on a latent space. Experimental results demonstrate that state-of-the-art learning systems on existing tasks reach only 12.01% of accuracy with raw input signals. Also, even in the oracle case with human-annotated descriptions, BERT contextual embedding achieves at most 28% of accuracy. Our proposed TrUSt boosts the model performance and reaches 13.94% performance. We also provide detailed analysis to pave the way for future research. TrUMAn is publicly available at:https://www.cmlab.csie.ntu.edu.tw/project/trope

Yueh-Cheng Liu, Yu-Kai Huang, Hung-Yueh Chiang et al.

Most 3D neural networks are trained from scratch owing to the lack of large-scale labeled 3D datasets. In this paper, we present a novel 3D pretraining method by leveraging 2D networks learned from rich 2D datasets. We propose the contrastive pixel-to-point knowledge transfer to effectively utilize the 2D information by mapping the pixel-level and point-level features into the same embedding space. Due to the heterogeneous nature between 2D and 3D networks, we introduce the back-projection function to align the features between 2D and 3D to make the transfer possible. Additionally, we devise an upsampling feature projection layer to increase the spatial resolution of high-level 2D feature maps, which enables learning fine-grained 3D representations. With a pretrained 2D network, the proposed pretraining process requires no additional 2D or 3D labeled data, further alleviating the expensive 3D data annotation cost. To the best of our knowledge, we are the first to exploit existing 2D trained weights to pretrain 3D deep neural networks. Our intensive experiments show that the 3D models pretrained with 2D knowledge boost the performances of 3D networks across various real-world 3D downstream tasks.

Yu-Kai Huang, Yueh-Cheng Liu, Tsung-Han Wu et al.

Dense depth estimation plays a key role in multiple applications such as robotics, 3D reconstruction, and augmented reality. While sparse signal, e.g., LiDAR and Radar, has been leveraged as guidance for enhancing dense depth estimation, the improvement is limited due to its low density and imbalanced distribution. To maximize the utility from the sparse source, we propose $S^3$ technique, which expands the depth value from sparse cues while estimating the confidence of expanded region. The proposed $S^3$ can be applied to various guided depth estimation approaches and trained end-to-end at different stages, including input, cost volume and output. Extensive experiments demonstrate the effectiveness, robustness, and flexibility of the $S^3$ technique on LiDAR and Radar signal.

Tung-I Chen, Yueh-Cheng Liu, Hung-Ting Su et al.

While recent progress has significantly boosted few-shot classification (FSC) performance, few-shot object detection (FSOD) remains challenging for modern learning systems. Existing FSOD systems follow FSC approaches, ignoring critical issues such as spatial variability and uncertain representations, and consequently result in low performance. Observing this, we propose a novel \textbf{Dual-Awareness Attention (DAnA)} mechanism that enables networks to adaptively interpret the given support images. DAnA transforms support images into \textbf{query-position-aware} (QPA) features, guiding detection networks precisely by assigning customized support information to each local region of the query. In addition, the proposed DAnA component is flexible and adaptable to multiple existing object detection frameworks. By adopting DAnA, conventional object detection networks, Faster R-CNN and RetinaNet, which are not designed explicitly for few-shot learning, reach state-of-the-art performance in FSOD tasks. In comparison with previous methods, our model significantly increases the performance by 47\% (+6.9 AP), showing remarkable ability under various evaluation settings.

Chen-Hsi Chang, Hung-Ting Su, Jui-heng Hsu et al.

The human ability of deep cognitive skills are crucial for the development of various real-world applications that process diverse and abundant user generated input. While recent progress of deep learning and natural language processing have enabled learning system to reach human performance on some benchmarks requiring shallow semantics, such human ability still remains challenging for even modern contextual embedding models, as pointed out by many recent studies. Existing machine comprehension datasets assume sentence-level input, lack of casual or motivational inferences, or could be answered with question-answer bias. Here, we present a challenging novel task, trope detection on films, in an effort to create a situation and behavior understanding for machines. Tropes are storytelling devices that are frequently used as ingredients in recipes for creative works. Comparing to existing movie tag prediction tasks, tropes are more sophisticated as they can vary widely, from a moral concept to a series of circumstances, and embedded with motivations and cause-and-effects. We introduce a new dataset, Tropes in Movie Synopses (TiMoS), with 5623 movie synopses and 95 different tropes collecting from a Wikipedia-style database, TVTropes. We present a multi-stream comprehension network (MulCom) leveraging multi-level attention of words, sentences, and role relations. Experimental result demonstrates that modern models including BERT contextual embedding, movie tag prediction systems, and relational networks, perform at most 37% of human performance (23.97/64.87) in terms of F1 score. Our MulCom outperforms all modern baselines, by 1.5 to 5.0 F1 score and 1.5 to 3.0 mean of average precision (mAP) score. We also provide a detailed analysis and human evaluation to pave ways for future research.

Hung-Ting Su, Chen-Hsi Chang, Po-Wei Shen et al.

We study a novel task, Video Question-Answer Generation (VQAG), for challenging Video Question Answering (Video QA) task in multimedia. Due to expensive data annotation costs, many widely used, large-scale Video QA datasets such as Video-QA, MSVD-QA and MSRVTT-QA are automatically annotated using Caption Question Generation (CapQG) which inputs captions instead of the video itself. As captions neither fully represent a video, nor are they always practically available, it is crucial to generate question-answer pairs based on a video via Video Question-Answer Generation (VQAG). Existing video-to-text (V2T) approaches, despite taking a video as the input, only generate a question alone. In this work, we propose a novel model Generator-Pretester Network that focuses on two components: (1) The Joint Question-Answer Generator (JQAG) which generates a question with its corresponding answer to allow Video Question "Answering" training. (2) The Pretester (PT) verifies a generated question by trying to answer it and checks the pretested answer with both the model's proposed answer and the ground truth answer. We evaluate our system with the only two available large-scale human-annotated Video QA datasets and achieves state-of-the-art question generation performances. Furthermore, using our generated QA pairs only on the Video QA task, we can surpass some supervised baselines. We apply our generated questions to Video QA applications and surpasses some supervised baselines using generated questions only. As a pre-training strategy, we outperform both CapQG and transfer learning approaches when employing semi-supervised (20%) or fully supervised learning with annotated data. These experimental results suggest the novel perspectives for Video QA training.

Kuang-Yu Jeng, Yueh-Cheng Liu, Zhe Yu Liu et al.

We proposed an end-to-end grasp detection network, Grasp Detection Network (GDN), cooperated with a novel coarse-to-fine (C2F) grasp representation design to detect diverse and accurate 6-DoF grasps based on point clouds. Compared to previous two-stage approaches which sample and evaluate multiple grasp candidates, our architecture is at least 20 times faster. It is also 8% and 40% more accurate in terms of the success rate in single object scenes and the complete rate in clutter scenes, respectively. Our method shows superior results among settings with different number of views and input points. Moreover, we propose a new AP-based metric which considers both rotation and transition errors, making it a more comprehensive evaluation tool for grasp detection models.

Yu-Kai Huang, Yueh-Cheng Liu, Tsung-Han Wu et al.

The performance of image based stereo estimation suffers from lighting variations, repetitive patterns and homogeneous appearance. Moreover, to achieve good performance, stereo supervision requires sufficient densely-labeled data, which are hard to obtain. In this work, we leverage small amount of data with very sparse but accurate disparity cues from LiDAR to bridge the gap. We propose a novel sparsity expansion technique to expand the sparse cues concerning RGB images for local feature enhancement. The feature enhancement method can be easily applied to any stereo estimation algorithms with cost volume at the test stage. Extensive experiments on stereo datasets demonstrate the effectiveness and robustness across different backbones on domain adaption and self-supervision scenario. Our sparsity expansion method outperforms previous methods in terms of disparity by more than 2 pixel error on KITTI Stereo 2012 and 3 pixel error on KITTI Stereo 2015. Our approach significantly boosts the existing state-of-the-art stereo algorithms with extremely sparse cues.

Yu-Siang Wang, Hung-Ting Su, Chen-Hsi Chang et al.

We introduce a novel task, Video Question Generation (Video QG). A Video QG model automatically generates questions given a video clip and its corresponding dialogues. Video QG requires a range of skills -- sentence comprehension, temporal relation, the interplay between vision and language, and the ability to ask meaningful questions. To address this, we propose a novel semantic rich cross-modal self-attention (SRCMSA) network to aggregate the multi-modal and diverse features. To be more precise, we enhance the video frames semantic by integrating the object-level information, and we jointly consider the cross-modal attention for the video question generation task. Excitingly, our proposed model remarkably improves the baseline from 7.58 to 14.48 in the BLEU-4 score on the TVQA dataset. Most of all, we arguably pave a novel path toward understanding the challenging video input and we provide detailed analysis in terms of diversity, which ushers the avenues for future investigations.