SynSpill: Improved Industrial Spill Detection With Synthetic Data

Large-scale Vision-Language Models (VLMs) have transformed general-purpose visual recognition through strong zero-shot capabilities. However, their performance degrades significantly in niche, safety-critical domains such as industrial spill detection, where hazardous events are rare, sensitive, and difficult to annotate. This scarcity -- driven by privacy concerns, data sensitivity, and the infrequency of real incidents -- renders conventional fine-tuning of detectors infeasible for most industrial settings. We address this challenge by introducing a scalable framework centered on a high-quality synthetic data generation pipeline. We demonstrate that this synthetic corpus enables effective Parameter-Efficient Fine-Tuning (PEFT) of VLMs and substantially boosts the performance of state-of-the-art object detectors such as YOLO and DETR. Notably, in the absence of synthetic data (SynSpill dataset), VLMs still generalize better to unseen spill scenarios than these detectors. When SynSpill is used, both VLMs and detectors achieve marked improvements, with their performance becoming comparable. Our results underscore that high-fidelity synthetic data is a powerful means to bridge the domain gap in safety-critical applications. The combination of synthetic generation and lightweight adaptation offers a cost-effective, scalable pathway for deploying vision systems in industrial environments where real data is scarce/impractical to obtain. Project Page: https://synspill.vercel.app

Re:Verse -- Can Your VLM Read a Manga?

Current Vision Language Models (VLMs) demonstrate a critical gap between surface-level recognition and deep narrative reasoning when processing sequential visual storytelling. Through a comprehensive investigation of manga narrative understanding, we reveal that while recent large multimodal models excel at individual panel interpretation, they systematically fail at temporal causality and cross-panel cohesion, core requirements for coherent story comprehension. We introduce a novel evaluation framework that combines fine-grained multimodal annotation, cross-modal embedding analysis, and retrieval-augmented assessment to systematically characterize these limitations. Our methodology includes (i) a rigorous annotation protocol linking visual elements to narrative structure through aligned light novel text, (ii) comprehensive evaluation across multiple reasoning paradigms, including direct inference and retrieval-augmented generation, and (iii) cross-modal similarity analysis revealing fundamental misalignments in current VLMs' joint representations. Applying this framework to Re:Zero manga across 11 chapters with 308 annotated panels, we conduct the first systematic study of long-form narrative understanding in VLMs through three core evaluation axes: generative storytelling, contextual dialogue grounding, and temporal reasoning. Our findings demonstrate that current models lack genuine story-level intelligence, struggling particularly with non-linear narratives, character consistency, and causal inference across extended sequences. This work establishes both the foundation and practical methodology for evaluating narrative intelligence, while providing actionable insights into the capability of deep sequential understanding of Discrete Visual Narratives beyond basic recognition in Multimodal Models.

LR0.FM: Low-Resolution Zero-shot Classification Benchmark For Foundation Models

Visual-language foundation Models (FMs) exhibit remarkable zero-shot generalization across diverse tasks, largely attributed to extensive pre-training on largescale datasets. However, their robustness on low-resolution/pixelated (LR) images, a common challenge in real-world scenarios, remains underexplored. We introduce LR0.FM, a comprehensive benchmark evaluating the impact of low resolution on the zero-shot classification performance of 10 FM(s) across 66 backbones and 15 datasets. We propose a novel metric, Weighted Aggregated Robustness, to address the limitations of existing metrics and better evaluate model performance across resolutions and datasets. Our key findings show that: (i) model size positively correlates with robustness to resolution degradation, (ii) pre-training dataset quality is more important than its size, and (iii) fine-tuned and higher resolution models are less robust against LR. Our analysis further reveals that the model makes semantically reasonable predictions at LR, and the lack of fine-grained details in input adversely impacts the model's initial layers more than the deeper layers. We use these insights and introduce a simple strategy, LR-TK0, to enhance the robustness of models without compromising their pre-trained weights. We demonstrate the effectiveness of LR-TK0 for robustness against low-resolution across several datasets and its generalization capability across backbones and other approaches. Code is available at https://github.com/shyammarjit/LR0.FM

PV-S3: Advancing Automatic Photovoltaic Defect Detection using Semi-Supervised Semantic Segmentation of Electroluminescence Images

Photovoltaic (PV) systems allow us to tap into all abundant solar energy, however they require regular maintenance for high efficiency and to prevent degradation. Traditional manual health check, using Electroluminescence (EL) imaging, is expensive and logistically challenging making automated defect detection essential. Current automation approaches require extensive manual expert labeling, which is time-consuming, expensive, and prone to errors. We propose PV-S3 (Photovoltaic-Semi Supervised Segmentation), a Semi-Supervised Learning approach for semantic segmentation of defects in EL images that reduces reliance on extensive labeling. PV-S3 is a Deep learning model trained using a few labeled images along with numerous unlabeled images. We introduce a novel Semi Cross-Entropy loss function to train PV-S3 which addresses the challenges specific to automated PV defect detection, such as diverse defect types and class imbalance. We evaluate PV-S3 on multiple datasets and demonstrate its effectiveness and adaptability. With merely 20% labeled samples, we achieve an absolute improvement of 9.7% in IoU, 29.9% in Precision, 12.75% in Recall, and 20.42% in F1-Score over prior state-of-the-art supervised method (which uses 100% labeled samples) on UCF-EL dataset (largest dataset available for semantic segmentation of EL images) showing improvement in performance while reducing the annotation costs by 80%.

Semi-supervised Active Learning for Video Action Detection

In this work, we focus on label efficient learning for video action detection. We develop a novel semi-supervised active learning approach which utilizes both labeled as well as unlabeled data along with informative sample selection for action detection. Video action detection requires spatio-temporal localization along with classification, which poses several challenges for both active learning informative sample selection as well as semi-supervised learning pseudo label generation. First, we propose NoiseAug, a simple augmentation strategy which effectively selects informative samples for video action detection. Next, we propose fft-attention, a novel technique based on high-pass filtering which enables effective utilization of pseudo label for SSL in video action detection by emphasizing on relevant activity region within a video. We evaluate the proposed approach on three different benchmark datasets, UCF-101-24, JHMDB-21, and Youtube-VOS. First, we demonstrate its effectiveness on video action detection where the proposed approach outperforms prior works in semi-supervised and weakly-supervised learning along with several baseline approaches in both UCF101-24 and JHMDB-21. Next, we also show its effectiveness on Youtube-VOS for video object segmentation demonstrating its generalization capability for other dense prediction tasks in videos.

GAMa: Cross-view Video Geo-localization

The existing work in cross-view geo-localization is based on images where a ground panorama is matched to an aerial image. In this work, we focus on ground videos instead of images which provides additional contextual cues which are important for this task. There are no existing datasets for this problem, therefore we propose GAMa dataset, a large-scale dataset with ground videos and corresponding aerial images. We also propose a novel approach to solve this problem. At clip-level, a short video clip is matched with corresponding aerial image and is later used to get video-level geo-localization of a long video. Moreover, we propose a hierarchical approach to further improve the clip-level geolocalization. It is a challenging dataset, unaligned and limited field of view, and our proposed method achieves a Top-1 recall rate of 19.4% and 45.1% @1.0mile. Code and dataset are available at following link: https://github.com/svyas23/GAMa.

Multi-modal Robustness Analysis Against Language and Visual Perturbations

Joint visual and language modeling on large-scale datasets has recently shown a good progress in multi-modal tasks when compared to single modal learning. However, robustness of these approaches against real-world perturbations has not been studied. In this work, we perform the first extensive robustness study of such models against various real-world perturbations focusing on video and language. We focus on text-to-video retrieval and propose two large-scale benchmark datasets, MSRVTT-P and YouCook2-P, which utilize 90 different visual and 35 different textual perturbations. The study reveals some interesting findings: 1) The studied models are more robust when text is perturbed versus when video is perturbed 2) The transformer text encoder is more robust on non-semantic changing text perturbations and visual perturbations compared to word embedding approaches. 3) Using two-branch encoders in isolation is typically more robust than when architectures use cross-attention. We hope this study will serve as a benchmark and guide future research in robust multimodal learning.

Large-scale Robustness Analysis of Video Action Recognition Models

We have seen a great progress in video action recognition in recent years. There are several models based on convolutional neural network (CNN) with some recent transformer based approaches which provide state-of-the-art performance on existing benchmark datasets. However, large-scale robustness has not been studied for these models which is a critical aspect for real-world applications. In this work we perform a large-scale robustness analysis of these existing models for video action recognition. We mainly focus on robustness against distribution shifts due to real-world perturbations instead of adversarial perturbations. We propose four different benchmark datasets, HMDB-51P, UCF-101P, Kinetics-400P, and SSv2P and study the robustness of six different state-of-the-art action recognition models against 90 different perturbations. The study reveals some interesting findings, 1) transformer based models are consistently more robust against most of the perturbations when compared with CNN based models, 2) Pretraining helps Transformer based models to be more robust to different perturbations than CNN based models, and 3) All of the studied models are robust to temporal perturbation on the Kinetics dataset, but not on SSv2; this suggests temporal information is much more important for action label prediction on SSv2 datasets than on the Kinetics dataset. We hope that this study will serve as a benchmark for future research in robust video action recognition. More details about the project are available at https://rose-ar.github.io/.

Video Action Detection: Analysing Limitations and Challenges

Beyond possessing large enough size to feed data hungry machines (eg, transformers), what attributes measure the quality of a dataset? Assuming that the definitions of such attributes do exist, how do we quantify among their relative existences? Our work attempts to explore these questions for video action detection. The task aims to spatio-temporally localize an actor and assign a relevant action class. We first analyze the existing datasets on video action detection and discuss their limitations. Next, we propose a new dataset, Multi Actor Multi Action (MAMA) which overcomes these limitations and is more suitable for real world applications. In addition, we perform a biasness study which analyzes a key property differentiating videos from static images: the temporal aspect. This reveals if the actions in these datasets really need the motion information of an actor, or whether they predict the occurrence of an action even by looking at a single frame. Finally, we investigate the widely held assumptions on the importance of temporal ordering: is temporal ordering important for detecting these actions? Such extreme experiments show existence of biases which have managed to creep into existing methods inspite of careful modeling.

LARNet: Latent Action Representation for Human Action Synthesis

We present LARNet, a novel end-to-end approach for generating human action videos. A joint generative modeling of appearance and dynamics to synthesize a video is very challenging and therefore recent works in video synthesis have proposed to decompose these two factors. However, these methods require a driving video to model the video dynamics. In this work, we propose a generative approach instead, which explicitly learns action dynamics in latent space avoiding the need of a driving video during inference. The generated action dynamics is integrated with the appearance using a recurrent hierarchical structure which induces motion at different scales to focus on both coarse as well as fine level action details. In addition, we propose a novel mix-adversarial loss function which aims at improving the temporal coherency of synthesized videos. We evaluate the proposed approach on four real-world human action datasets demonstrating the effectiveness of the proposed approach in generating human actions. Code available at https://github.com/aayushjr/larnet.